CHAPTER 4 (Part 2)

169. Stewart makes a mistake similar to that noted above when he expounds his views in another way:

170. Anyone studying this passage can easily recognise the uncertainty and hesitancy of its author. Having no clear proof for his system, he tries to support it with a line of argument full of approximations intended to suggest a link between ideas where, in fact, there is none.
Look at the final words of the passage just quoted. The statement, 'This quality constitutes its nominal essence but not its real essence', implies the presence of two essences rather than one, and therefore admits more than he intends to deny.

171. But I shall not be over-critical of Stewart's use of words. Rather, let me ask whether by nominal essence he means a word. This would seem to follow from what he says in other passages and from the aim of his argument which is to show that universal ideas are nothing.
If by the term nominal essence he does not mean a mere word but something more, his whole argument is invalid. In that case, the general terms would express something objective, and would not be mere words.

172. In the passage I quoted, Stewart himself admits this. He calls nominal essence a quality actually possessed by the individual, and adds:

If this quality were nothing, the individual could neither possess it nor receive from it the generic name. Moreover, Stewart himself attributes to the human spirit the faculty of thinking of a quality of an individual without conceiving the other qualities which go to make up the individual. He says:
The classification of different objects supposes a power of attending to some of their qualities or attributes, without attending to the rest.(117)
Thus he accepts that 1. the single characteristics of individuals are real; 2. we have a faculty for considering them on their own, separate from the individuals themselves (considering them on their own is the same thing as considering them while prescinding from everything with which they coexist); 3. when our spirit considers these qualities on their own and in isolation, it has a true object before itself because these qualities are true.
Let us consider the qualities of bodies: colour, taste, smell, sound, extension, hardness, fluidity, and so on. Setting aside for the moment the question of the existence of bodies and assuming with Stewart himself that they are real, we have here, according to Stewart's own principles, as many qualities, as many mental objects, as there are bodies.

The names therefore of these qualities, that is, the words colour, taste, and so on, which are all abstract nouns, also express something effective. They are not merely names, but have something which actually corresponds to them, that is, they have these qualities, whatever these may be in things. If abstract words such as the colour, taste, etc. of bodies are not merely names but have something beyond what they signify, it follows that common nouns and appellatives such as coloured, tasty, and so on, body, man, and so on, also have some effective meaning. In fact, they are simply nouns meaning that which has colour, taste, and so on, that which has corporality, humanity and so on. Common nouns, therefore, are not mere words without any corresponding object but, in accordance with Stewart's own principles, signify some object proper to them.

173. At this point, Stewart may reply: 'I cannot deny that abstracts and common nouns indicate something. If I have denied it anywhere, this was merely an inaccurate expression. However, I maintain what they indicate is merely "the particular quality or a group of qualities by which an individual resembles other individuals." Consequently, it is in no way universal, but entirely particular. This quality is only in the individuals where it is always individual.'

I shall certainly not bring up again Plato's question: do abstract qualities have an existence outside spirit and distinct from entia themselves? This would contribute nothing to my purpose. I am perfectly willing to grant Mr. Stewart the right to his view that the qualities to which we are referring do not have an existence outside a spirit except in the individuals themselves. But he likewise accepted, with me, that our spirit can and does consider them in isolation from individuals, and as if these qualities existed alone. This is a fact which cannot be disputed.

From this I conclude: our spirit, if it considers qualities in isolation from individuals, has an immediate, universal object for its focus, that is, a quality distinct from an individual is a universal object completely independent of the word used to express it.

174. If I show conclusively the truth of this last assertion, I believe that these consequences will follow: 1. our spirit can have a universal object; 2. it can give this object a name, and consequently 3. there are nouns which express universal ideas. These are not mere words devoid of meaning nor words for which, by a blind habit, we substitute certain individuals.
When I say that a quality, considered in the way in which our spirit can view it, that is, apart from an ens, is universal, I simply mean that I can conceive it in an indefinite number of individuals. That it can be thought in an indefinite number of individuals, or be universal or common, is synonymous in the sense normally given to these words.
The particular being corresponds to the universal being of a quality. This simply means that it cannot be thought of as common to a number of individuals, but as fixed and proper to one individual. The individuality of the ens to which it is applied makes the common quality particular. This explains why qualities remain common for us as long as we do not think of them as existing in determinate individuals. In other words, we think of them in such a way that we retain the option of imagining them as connected with any individual, until we have linked them to one individual. Once they are linked to this individual, they too are individualised by it and are called particular. In this case, the whiteness, size, and so on, of one body is not the whiteness or size of another body.

If a quality is particular only insofar as it exists in an individual and if, as I have said, our spirit has the faculty of considering it without considering the individual to which it belongs (which Stewart himself concedes), I conclude that our spirit has the faculty of considering it as merely possible without thinking that it has a real existence in some individual. Dr. Reid calls this simple apprehension; Professor Stewart seems to call it conception. If this cannot be disputed, and our spirit can think of whiteness not as something really existing but as merely possible, I maintain that in such a case the object of our spirit is universal in the sense in which the Ideologues use the word. This whiteness is not linked to any individual; it is whiteness which we conceive as capable of being received by an indefinite number of individuals. We think it in such a way that, if we had the faculty to create, we would be able to realise it, from the idea we had of it, in an indefinite number of white bodies.

175. This whiteness conceived by our mind is not therefore merely a name, as Stewart would seem to maintain, nor is it any of the whitenesses we have seen really existing in the white bodies under our gaze.
It is not any of these really existing whitenesses all of which are particular whitenesses, as I have said, in the individual in such a way that they cannot be transferred from one individual to another nor joined with a number of individuals even by thought.
How could I conceive a way of transferring the whiteness of one white body to another without depriving the former of its whiteness? The white body in question either has a white surface and the rest of a different colour, or it is completely white like chalk which, as friable, leaves its own whiteness on bodies it touches. Let us now consider the difference between making bodies white by means of the whiteness really existing in an individual, and making them white by means of the idea of general whiteness which is, I maintain, in our mind.

1. In the first place, a body cannot transfer its own whiteness to another, however white it is, unless it is friable. But at the same time its parts are so hard that they cannot easily release the tiny fragments that will cover and whiten the surface of the other body. On the other hand, anyone who has the power to create, creates bodies endowed with whiteness by giving them this quality, but does this by drawing the whiteness only from the idea of whiteness present in his spirit. This idea has no need to be friable nor possess any other quality in order to be communicated to bodies.

2. If a white body which will pass some of its whiteness to another body has only a slight surface area of white, it will deprive itself of a slight covering of colour when it whitens the other body. On the other hand, the intelligent spirit to which I am referring is able to create white bodies at a stroke, as it conceives them possible, without reducing or destroying the notion it has of general whiteness.

3. When the colouring body, such as chalk, is friable and completely white, it cannot make another body white without losing a thin white coating which is attached to the whitened body. In losing this thin coating, the whitening body, although still seen as white, does not present the same previous whiteness. The white surface previously seen has been transferred to the other body; the first body now reveals another surface, white like the first, but not the first.

We can infer, therefore, that it is not strictly accurate to say that the whiteness really existing in an individual is transferred to another. When a white body whitens another body upon contact, it is not the same whiteness communicated to two bodies, nor one whiteness transferred from one body to another. The first body is an amalgam of many particles or tiny white bodies which detach themselves or are taken from the walls of the first body and settle on the walls of the second. In this way they whiten the other body but do not communicate their own whiteness; the particles simply change position. The second body, despite appearances, does not in fact change colour.
Indeed the whiteness really existing in individuals is so particular to them that it is wholly incommunicable. Although the bodies possessing it may be minced and pulverised, and the powder derived from them change position, nevertheless, the whiteness alone is never transferred in exactly the same way from one body to another.

On the other hand, if we were to imagine a spirit able to create white bodies, we certainly cannot imagine it as removing and erasing the real whiteness of the bodies and transferring it to other bodies which it wished to produce. That particular whiteness is incommunicable. However, we can think and imagine the spirit as giving existence to particular whitenesses on the basis of the standard, general whiteness which it contemplates in its mind.

4. Finally, even if we supposed that a white body communicated its whiteness to another, it could not communicate it to an infinite number of bodies. In this selfcommunication, it would become increasingly weaker by losing a thin coating of its substance to each body that it whitened. Eventually, it would vanish completely.

On the other hand, the quality of whiteness conceived in a universal manner by the intelligent mind, makes this spirit, which we imagine endowed with creative power, capable of creating an infinite series of white bodies without any diminution to the whiteness, or without its becoming less apt for renewed realisation in countless other bodies.
It is, therefore, this quality of whiteness, not the particular quality received in an individual, which enables a creative spirit to realise whiteness in an indefinite number of white bodies. The particular quality is of its nature incommunicable to other individuals.

176. Nor can we say that a spirit, which we imagine endowed with the faculty to create, imparts whiteness to the bodies it creates without needing to possess the idea of whiteness in addition to creative force. Such a force does not determine the creator to create bodies of one colour rather than another. This power cannot be thought as determined to create anything unless the understanding puts forward the objects which it creates.
It would also be unreasonable to reply that, in the hypothesis of a creating ens, we can no longer propose any argument because the idea of creation transcends both the way we conceive mentally and the rules governing our thought. The idea of a creative spirit was only introduced into the discussion to make the matter clearer; it does not mean that my argument relies on such an assumption. All that is needed for my argument is to take the example of a man who imagines as many white bodies as he likes. I can also ask him whether the whiteness he imagines is the whiteness he sees in individuals. It seems obvious to me that it is not, just as it is obvious that the whiteness which a creating ens would transmit to created bodies is not. The whiteness seen in individuals is inseparable from them; it is individual and incommunicable by nature. But the whiteness which our imagination bestows on possible bodies is indefinitely communicable.

The obvious reason why these two whitenesses are not identical is this. On the one hand, we know we perceive the whiteness of the white bodies each time we see them, and we realise that the whiteness related to those bodies is inseparable from them. On the other hand, we are aware that we can imagine many other white bodies similar to those we have seen.
Let us assume that we imagine all the white bodies we have seen during our life. Now, using our imagination, we could add a similar number of possible bodies as white as those we have seen. Is the whiteness of this imagined and thought-out array of bodies, together with that of the bodies we have actually seen, the whiteness of the seen bodies or some other whiteness? It cannot be the whiteness of the actually seen bodies because this is individual, and we have already assumed that all the bodies we have seen are present to our inner gaze. In addition to all the whiteness it has seen, our spirit can endlessly conceive another whiteness and a whiteness which is not real but purely imaginary, or rather purely thought. In fact, we are dealing here - and this must be noted - solely with the object of thought.

177. If our spirit were restricted to conceiving or recalling the whiteness it had seen in bodies, its only faculty, apart from sense, would be that of recalling phantasms. But all agree that in addition we possess conception and imagination. If we consider only the imagination, we see that we are able at will to multiply infinitely entia similar to those we have seen. It is this faculty which needs explanation. It cannot, however, be explained in any way by assuming, with Stewart, that our spirit is bereft of universal ideas, that is, of ideas which stand for qualities in isolation from individuals and without the faculty of attributing these qualities to an indefinite number of possible or thinkable individuals [App., no. 7].

Now, if our spirit can conceive whiteness only in an indefinite number of possible individuals and is not obliged to think also of the determinate individual in which it exists, and if this possible whiteness is not the whiteness existing in single entities which we can see (since the definition of whiteness is whiteness as thought, after removal from the thought of the individuals to which it really belongs) it follows, I maintain, that this whiteness is not by any means a mere name. What I said earlier about common nouns would seem a sufficient demonstration of this. Nevertheless, I think it worthwhile proving it once more, granted the propensity of modern philosophy to nominalism.

If whiteness, as thought and not existent in any of the entia we see, were merely a name, we would be doing precisely nothing whenever we imagined white objects but did not name them. But no-one will allow himself to be persuaded that his mind, in imagining things he has never seen, experienced or named individually, is doing nothing. We have all found relief at times from our woes in pleasant, though vain and imaginary images. At times we all enjoy pursuing pleasant dreams presented at happier moments by the wondrous power of intellective imagination when we are fully awake. The lover will never be persuaded that his daydreams are merely pleasant illusions, without reality, and non-existent in his spirit or soul. The poet will not be persuaded that his lovely verses are vain and wasted words when he expresses in them individual objects he has never seen, touched and felt. If these words, which have no reference to actually existing objects, are empty, meaningless sounds, how can a sublime poet enchant all his contemporaries by his almost divine art and astound later ages? Does he alone have the gift of discovering magical, powerful sounds, devoid of meaning? Where does he derive such sounds? What God inspires him to utter them? What spirit moves his lips mechanically to utter them? Before uttering them, has he no concept, no thought, no imaginative representation present in his spirit as his whole song deserts the sensible world and takes flight beyond the afflictions of these individual things to range over the interminable fields of endless imagination?

Finally, what would a person, known for his new discoveries and enterprises, say to a philosopher who dryly insisted: 'Listen, you cannot conceive anything except already existing individuals. You are wasting your time if you are thinking of helping the world with beautiful, new discoveries, or some original writing or some magnanimous enterprise. When you think about such things which do not yet exist, you are just like the idiot who does nothing and thinks of nothing. When you talk about these things, you are nothing more than a trickster. In fact, you are not even that, because your words and sounds are purely empty, like the noise of a collection of stones rubbed or knocked against each other. They express nothing existent, no particular individual, without which there is no thought.'

Anyone stipulating that we have no ideas of the individual qualities of entia unless they are observed in the individuals we have seen, and that such qualities, when considered in isolation from the individuals as being merely possible, are mere names - this is Stewart's case - unwittingly and unwillingly renounces and disowns all arts and sciences. He has no means of explaining intellectual imagination. According to such philosophers, human beings can have only the threadbare remembrance of things they have seen (and even this remains unexplained) [App., no. 8]; they cannot imagine possible entia. This blocks the wellspring of all rational and human activity, which has its source in the power to carry out and obtain possible, future good. To imagine possible things, their qualities must first be considered in the mind as possible, that is, as qualities shareable by indefinite, but as yet non-subsistent entia.

178. What I have been saying so far has particular force against Stewart's system thanks to his dependence upon Dr. Reid's principles on the nature of ideas.
Reid denies the existence of ideas considered as an intermediary between the real objects of the spirit and the spirit itself. Locke, like the ancients, distinguished between ideas and things. He considered ideas, not things, as the proximate term of the intellect. Reid, however, did not want any intermediary between the real, perceived ens and the perceiving spirit. This is also Stewart's view.
Now, as far as individuals are concerned, the real object truly exists because real individuals exist. However, Reid's system had no way of explaining universal ideas which have no existence outside the spirit. As a result, Stewart decided to deny them completely; for him, they are names, and nothing more [App., no. 9].

179. I shall not enquire here if this part of Reid's system is true or false. I have already dealt with it. Nor shall I examine if Stewart understood it correctly, and if the refusal to accept universals completely while supposing that mere names may take their place is an inevitable consequence of the system.
I need only point out that Stewart thought he was obliged to do so by the strict necessity of the system. His acceptance of the principle that there are no intermediate ideas between objects really existing outside us and ourselves who perceive them led him to deny completely the existence of universal ideas in which the spirit has no really existing object.

But I have proved that: 1. names are not sufficient to explain the act whereby the spirit imagines possible entia, and in greater number than all the individuals it perceives by the senses; 2. the ideas of qualities perceived in individuals themselves (insofar as qualities adhere to them) are not sufficient either; 3. it is moreover necessary for our mind to conceive these qualities in themselves, that is, in isolation from the individuals and hence simply as possible. It seems obvious, therefore, that Stewart's system is defective and inadequate; it provides no explanation of this final mode of conception by which universal ideas are formed and are present.

180. There are still a number of reflections to be made on the passage quoted from Stewart.

First, the reader should consider the phrase Stewart uses when he states what he understands by the essence of an individual. He says:

The odd thing about this passage is that nobody can disagree with this definition. I am quite certain that Plato himself would have had nothing to add to it. This means that Stewart's passage leaves untouched the question it was meant to deal with.

181. It is true that in the passage he does not mention the words universals and general ideas and similar terms. What I maintain, however, is that the passage contains the meaning of those very words which have been studiously avoided. Consequently, by the use of such terms, universals have not been banished from metaphysics although it would seem that fear has prevented the use of their proper names.

182. To see how this comes about, I would ask the reader how he would interpret our philosopher's phrase, 'the quality by which an individual resembles other individuals.'
He may answer that he does not feel it necessary to enquire about the likeness one thing has to another. Everybody understands the expression: one individual is like another. I too believe that everybody can understand it, and I believe that it can be easily defined.

'Two or more individuals resemble one another,' obviously states less than: 'Two or more individuals are the same.' One cannot say that a number of individuals are the same unless they are the same in all their parts and qualities. For them to be similar, on the other hand, all that is needed is that they be the same in some particular quality. There is no similarity between a number of objects except under some aspect they have an equal, common quality. But I do not wish to break off at this point to work out what follows from this relative to the nature of the identical or common quality. I would rather point out that I can never know the likeness or identical nature of a number of objects if in my mind I have only their individual idea or the idea of their individual qualities. In fact, the qualities of two objects, insofar as they are individual, that is, adhering to an individual, cannot be compared with each other in any way because the qualities found in one individual are found in a different place from those in some other individual. Two things to be compared can never be put together for comparison as long as they are found in different places. Comparing a number of things or qualities to discover how they are alike and how they differ requires an intellective spirit which possesses not only the faculty of perceiving them individually, but that of mentally isolating them(118) from the individuals and linking them together. Through this comparison, we discover what they have in common and what is proper to them.

The surveyor wishes to discover whether two triangles are equal; he mentally superimposes one upon the other and observes if they meet exactly. Similarly, the carpenter superimposes one table upon another when he needs to see whether two tables are the same size. Yet the carpenter's action is quite different from the surveyor's. Note that merely placing the tables close up to one another would be useless. Mere physical collocation would not enable the carpenter to see if the two tables were equal unless he possessed in addition within himself an intelligent spirit, capable of conceiving them interpenetrating each other, that is, both occupying the same space. If the spirit wishes to compare two lines, it must put one line in the place of the other. If it wishes to compare two surfaces, it has to conceive them one inside the other. If it wishes to compare two solids, it is obliged to conceive them as mingled in all their dimensions. Thus the spirit sees if they are equal or unequal, and which of the two is greater, which smaller. However close and coherent the two physical solids become, they still remain apart and are not, therefore, truly compared with each other. One exists and has no relationship whatever to the existence of the other.

Someone may say: If, in placing side by side two solids to see which is greater, the carpenter does not produce any comparison except in his spirit, why does he bother to place them side by side? He does so, not in order to make a comparison outside his spirit, but because, by this external action, he enables his spirit, and his imagination also, to carry out a true interior comparison. Moreover, there seems no possible doubt about this to anyone who tries to discover how our spirit makes a comparison between two or more things.

183. I must simply point out that what I have said about bodies and extension in the example also applies to any two individuals whatsoever. Two individuals can never be entirely intermingled. As individuals, they have two distinct, independent existences. It can therefore be affirmed that if there were only individuals, they could never be compared because they could not be lodged in the same place or, to put it in a more general way, in one and the same existence.

184. What does the spirit need to enable it to compare two or more individuals and ascertain how they are equal and unequal, how they are alike and how they differ? According to Stewart, and Reid before him, the spirit has only strictly individual ideas which do not differ from the individuals themselves which are enveloped by thought. However, these individual ideas are insufficient to set up a comparison, just as the individuals, from which these ideas do not differ relative to their distinctiveness and independence, are insufficient. In fact, the idea of a quality would cease to be individual if this quality which we think about could, by virtue of our thought, be passed on from one individual to another. A quality is particular or individual solely when it is conceived as adhering to a single individual. Thus, just as there is no comparison established between two individuals in isolation from the spirit which compares them, so there is no comparison between two individual ideas, one of which can never (precisely on the assumption that they are merely individuals) be confused or identified with the other. For the spirit to find that two individuals are similar or dissimilar, it must also possess some universal idea in addition to the individual ideas as I shall explain.

185. The issue here is to ascertain how two white surfaces are like each other, where one is whiter than the other.
Neither the surfaces themselves nor the individual whiteness of the surfaces can be transferred from one to the other. If it were possible, the two whitenesses would produce a third, which would still not furnish the intended comparison between the first two whitenesses. Nor can the idea of the individual whiteness of one surface be compared with the idea of the individual whiteness of the other surface without some intermediary aid. When I say individual whiteness, I mean whiteness which has such a proper existence that it cannot be externalised, nor transferred to another surface, nor take any other into itself; it is alien to any other, unaware of it, and excludes it. The means by which our spirit is able to compare the two forms of whiteness of which we are speaking has to be a potency through which we have a universal notion of whiteness, not the mere sight of some individual, existing whiteness. Only to the universal notion can we directly compare the individual whitenesses perceived by the senses and see to what extent they share in the notion of white.

Indeed, let us suppose that we have formed in our mind the idea of universal whiteness (how does not matter for the moment). This whiteness has not been received in any existing individual, but stands on its own so that we view it as capable of being actualised in an infinite number of individuals.
Such an idea, untrammelled, so to speak, by the individual in our spirit, is by nature a type, an example, a rule we use to judge rapidly the resemblance of the sensible individuals which come under our gaze. We do this as follows. As we see a white surface, we have in our spirit 1. the perception of the surface in question; 2. the universal idea of possible whiteness. We then compare this second whiteness with the first, and thus judge it. Such a comparison is possible because the universal idea of whiteness, unrestricted to any individual, can be conceived by us in all possible individuals and, therefore, in the individual whose whiteness we intend to judge. Thus the individual, felt whiteness and the mentally conceived universal whiteness become involved with one another, that is, are found together without being confused because it is impossible to confuse the general with the particular. Nevertheless, the particular is received in the general where it can be seen without losing the determination which makes it particular.

When we make a similar judgment about another surface, we have two individual surfaces, both judged to a certain degree of white.
Consequently, in accordance with the axiom that two things similar to a third are similar to each other, we discover the similarity of the two white surfaces.
In order to discover whether two or more individuals resemble one another we must suppose that in our spirit there is a common type or example of that quality which makes those individuals similar. This type or example, then, is simply the same quality considered by our spirit, but in isolation from all individuals and consequently in a universal manner. In short, it is the same quality, no longer as really existing but as possible, and capable of being received in an indefinite number of individuals.

186. If anyone finds this method of explaining how we discover the similarities between things unsatisfactory, I would be pleased were he to offer a more satisfying explanation.
It seems odd to me, though, that in a study of the difficulty of the nature of universal ideas formed by the spirit, anyone is happy to state that such an idea is merely 'the particular quality which renders an individual similar to other individuals of the same class.' He shows in this way that it is useless and superfluous to explain how the similarities in individuals are known. If it is useless to explain how the spirit comes to know similarities and differences, it is equally useless to undertake a study of universal ideas. These are not two questions, but a single, identical question expressed in different words. As I said, I cannot conceive how it is possible to make a judgment on equality or similarity between two objects, without some common yardstick which, precisely because it is common, cannot be individual, but must be universal.

187. If these yardsticks, these common qualities, these universals (such words are synonymous in our discussion) cannot be properly understood, or perhaps contain something mysterious and recondite, it does not follow that they can be rejected. That unfortunately is what human philosophy tends to do. If something cannot be understood, or is found mysterious, philosophy is quite prepared to deny it or maintain that it is an illusion, a dream of our backward, ancient world. At best, it describes the difficulty as inexplicable. Each thinker judges human ingenuity on the basis of his own mental forces, and makes this the limit of his philosophical modesty.

However, whatever present or past writers may say, the true lover of wisdom will think it imperative never to deny the existence of anything well-proven simply because he cannot grasp it. He prefers rather to confess ingenuously that he still does not understand its nature rather than declare that it is unintelligible and outside the scope of human investigation. Such statements can be left to the encyclopedists.

188. I should like to make one further remark on the passage from Stewart to which I referred. He writes:
It is this quality, therefore, which may be said to be essential to an individual in its classification under a particular genus. However, as every classification is to a certain degree arbitrary, we cannot conclude that this generic quality is more essential to the individual's existence than a host of other accidental qualities.
In setting out to explain a fact which is the subject of great controversy, it is necessary, I think, to avoid the use of ambiguous terms which may generate doubt and uncertainty. Great care is needed also to ensure that ideas relating to all the words are scrupulously examined. Stewart, however, does not seem to have examined the idea of classification in a genus. If he had done, he would easily have seen that this classification can be carried out only by means of a common idea, that is, by means of the quality which makes the individuals resemble one another precisely because it is common to them. But Stewart, in his use of the word classification, falls into the logical error of petitio principii just as he did when using the word likeness. To explain a fact, he assumes the fact as explained. He claims that there is no difficulty in classifying objects and uncovering their likeness. But this is the very difficulty that thinkers were seeking to explain. In other words, he has defined something by itself, idem per idem.

189. The expression he uses, 'as every classification is to a certain degree arbitrary', is also odd. Is this a strictly philosophical expression?
My objection would run as follows: 'When you state that every classification is to a certain degree arbitrary, you clearly admit that it is not arbitrary in all respects. So why not examine what is arbitrary and what is not in these classifications? Your reluctance to carry out this inquiry entitles your reader to suspect that the non-arbitrary aspect of classification is precisely the nub of the question. He will say that classifications are only based upon qualities which make things similar, or upon qualities they have in common (the two expressions are synonymous). Consequently, we have to accept that not all classifications called genera and species are arbitrary. The common qualities are neither arbitrary, nor mere names, but qualities actually existing in individuals. Your perfunctory admission that the formation of such classes of possible individuals (classes designated as genera and species) involves a non-arbitrary element - but one that is necessary and real - is tantamount to casting doubt on your whole system. It enables careful readers to discover for themselves, by a process of reasoning, the ruinous defect of the system.'

190. Lastly, I would point out that, in the short passage I have quoted, Stewart runs together and confuses two entirely different questions.
First: Are there universal ideas in the human spirit, that is, does a person think of the common qualities of things as merely possible?
Second: What are these universal ideas or common qualities of things outside the human spirit?
These two questions should not be confused and treated as one; the second, as I shall subsequently maintain, must be broken down into others.

191. The question about the existence of a common quality outside the spirit is an inquiry which has no bearing on my argument.

We all agree that outside the mind common qualities of themselves have no separate existence. They do not really exist unless they are made individual, that is, in the individuals to which they belong. However, despite this agreement, we still need to answer the first question: do common qualities exist in our spirit? Are they an object of knowledge?
This final inquiry, of course, is certainly very easy and obvious, provided our spirit is free from the sophistry into which present-day masters have led us. Over-reliance on their intellectual ability causes them to lay traps to catch their fellows rather than attain the truth.

192. Good sense tells us that the qualities of things are objects of thought not only as individual, but as common. A moment's reflection upon self makes us realise that 1. our spirit can know these qualities as they are present in this or that individual (knowledge of individual qualities); 2. we can consider them prescinding from the individual in which they are seen and perceived (we think of them as common); 3. consequently, we can grasp that certain qualities are simultaneously shared by a number of individuals and can also be shared by an infinite number of possible individuals. If this were not the case, I would be quite unable to think and express in words what I have thought here.

193. I do not wish to move on without first pointing out how Stewart introduces the other difficulty about common qualities considered as the essences of things although it has not the slightest bearing on his argument. Other philosophers regularly do the same:(119) they confuse the platonic question with our present problem. Moreover, they present it in an extremely inaccurate, false manner.

Where, for instance, did Stewart find that ancient philosophers made the essences of things consist in their common, universal qualities? As far as I can see, they too distinguished common qualities into essential and accidental qualities; and they formed genera and species from both. In fact, every common quality, accidental or essential, can be the basis for the formation of a genus or a species. But, if I speak about species of white and of black men, or like Aristotle, classify animals by the number of their legs, I have taken as the basis of those species an accidental quality, that is, the colours white or black and the number of legs. It seems to me that this twofold way of forming genera and species has always been distinct. Moreover the property of containing the true essence of individuals was attributed to genera and species formed in the first way, that is, on the basis of an essential quality.(120) On the other hand, genera and species formed in the second way on the basis of accidental qualities, were never thought to contain the true essence of individuals, but only their essence in so far as they belonged to that accidental, arbitrary species.

194. This second species could be called nominal(121) in a rather improper sense, but Stewart can never, strictly speaking, call the first nominal. The second species has an arbitrary element in it: if the issue is the formation of species based upon common accidental qualities, I decide which accidental quality to select. In the first sort of species, based upon an essential quality, there is nothing arbitrary. The essence of a determinate ens is unique, and must be used either to form the genus or abandon it.

However, as I said, the term 'nominal' would not be correct either. Calling this quality a nominal essence could lead people to think that it was merely a name. This, as I have shown, is false. The common qualities of things, whether accidental or essential, have an existence at least as objects of our spirit.

195. Stewart finds it impossible to imagine that an object can exist in our spirit without there being some corresponding thing outside. Consequently, after expounding the views of the two schools, the Realists and Nominalists, and coming down on the side of the Nominalists, he goes on to mention the intermediate group of Conceptualists and frankly admits that he is unable to form a clear idea of their teaching. He then sets about speculating on or rather guessing at the nature of their hypothesis.

He finds it only in two propositions which he puts forward as follows:

Immediately after this, he adds:

196. The Conceptualists agreed with the Nominalists in denying the subsistence of universal essences in themselves, but did not agree with them in denying the existence of universal ideas in the mind [App., no. 10]. In other words, they admitted that our spirit certainly did have universal concepts but that these concepts or ideas had no real existence outside our spirit. In short, they were ideas produced by the spirit at the moment of particular perceptions of things received by the senses.

197. In this system, the spirit came to possess 1. particular perceptions; 2. the faculty of working upon particular perceptions and adding to them a new form making them universal.
In fact, our spirit has the power to carry out operations on its own ideas and alter their form.(124) All the idols of our fantasy are simply aspects of the activity of our spirit which, as such, that is, in their very form, have no reality outside the spirit. They are tasks undertaken by the spirit through the operation of sensible things upon sensations and ideas.

198. On the other hand, Stewart considers the question of the need for language as the essential element in characterising the opinions of the three schools of which we are speaking, Realists, Conceptualists and Nominalists. He considers this question as an essential part of the problem of universals to which these philosophers offer different solutions. He assumes that the Realists are obliged by their system to think that words are not necessary to conceive universals. After stating that the difference between individuals and genera relative to the use of language consists in our capacity for reasoning about individuals without language, but not about genera, he goes on to say:

199. However, the question of the need for language is completely independent of the question which divided the three philosophical schools. Confusing them can only serve to make the main question extremely difficult and inextricable.
Although I have no intention whatsoever of being a Nominalist, I am convinced, on the other hand, of the need for words if we are to be drawn from the first to reflect upon universals (Theodicy, 100-102).

There is a great difference between supposing that universals are mere names to which neither things nor ideas correspond, and supposing that universals are things existing in themselves or, at least, ideas existing in our spirit, but in such a way that we can neither know these things nor possess these ideas for the first time without the aid of words.
Both those who hold the first opinion and those who hold either of the second opinions (that is, Nominalists as well as Realists and Conceptualists) may consider language necessary to enable mankind to start to think of universals. On this issue, there is only one difference between them. Nominalists are bound to believe language necessary. The other two schools may think so, but are not bound to do so in virtue of their opinion about the nature of universals. Nominalists are bound to consider language necessary to attain universals because, according to them, universals are only words. On the other hand, Conceptualists and Realists, if they consider language necessary, hold this opinion not by considering words as taking the place of ideas,(126) but as a suitable and necessary means of rousing and directing the attention of our spirit (which in itself is inert) towards the common qualities or of carrying out the operations on our perceptions by which we have universals.

200. I have purposely reserved for the end of these observations on Stewart's teaching on universals the author's most powerful passage in support of his case. This makes it easier to grasp its force and the force of the refutation which I am about to set out. Moreover, the ideas which I have been explaining while studying other passages of our philosopher could be helpful to readers.
In the following passage, Stewart attempts to determine how we can reason about universal truths with the help of words only and without permanently adding ideas to words. I shall quote the whole passage, despite its length, so that there can be no suspicion of my distorting his views. According to Stewart, these are the steps by which we attain universal truths.

He thinks, therefore, that we can reason about general truths by simply placing before ourselves either individuals or words. He explains his idea as follows:

201. Here I feel I must interrupt Mr. Stewart's argument for a moment to ask him what exactly he is seeking to achieve by it.
He is trying, he says, to explain universal truths, that is, trying to explain the formation of genera and species. In this case, I feel I have to ask him to examine the following phrases used by him in his argument: The circumstances common to the genus - the circumstances whereby the subject of our argument resembles the other individuals of the same genus. These two phrases (I want to concentrate solely upon them) naturally presuppose formed genera and presuppose that we make use of them. But how can he introduce already established genera and species into an argument whose aim is precisely to explain the formation of genera and species? This is another example of a blatant petitio principii.

202. Stewart goes on:

I raise no objections to this statement: it is in fact correct. However, it remains to be seen whether, if it is correct, universals are to be considered simply as names or whether the opposite is the case. This will become clear as soon as our author's theory is explained.
Stewart's own comment about the two ways he established for the attainment of general truths, is I think, attractive and perspicuous, and seems to throw light on the subject.

This is perfectly true; this excellent remark reduces the two methods of attaining universal truths to only one. The human spirit attains its aim by means of signs, which may exist in two forms: signs exhibiting similarity with the thing signified, and purely arbitrary signs totally without any similarity to what is signified. The picture depicting things has the first form, the letters of our alphabet have the second form. Geometry, which uses figures, has signs similar to the thing signified; algebra, which uses letters, has signs totally unconnected with what is signified.

203. I maintain that the very use of such signs implies the existence of universals and that there is no way, as Stewart claims, in which these signs are of themselves sufficient to explain how we use universal truths.
According to Stewart these signs enable us to attain general truths. But if these truths were nothing and did not differ from the signs themselves, what sense is there in speaking in this way? It would mean: 'By means of signs we attain signs; and not different signs but precisely those we make use of.' This is an odd, useless kind of philosophy, seemingly of little importance. My question to Stewart - and to anyone else in their right mind - is this: Surely the mere word sign directs our mind immediately to the thing signified? Could anyone possibly grasp the meaning of the word sign or things signified without immediately conceiving the idea of both of these things, related in such a way that one inescapably calls for the other?

204. Consequently, signs on their own cannot explain how we attain universal truths unless these truths are, in fact, something.
Saying that these signs direct our spirit to thinking of individuals is still not sufficient, as I have shown (cf. 198-199).
And, indeed, when I am told that a sign must focus my attention on a single, determinate individual, I understand perhaps how, for such a requirement, I need only to grasp two things, the sign and the individual signified. However, if I am told that a sign must lead me to think not about a single, determinate individual but about any individual whatsoever of a given genus or species (and no other individual outside this genus or species), I cannot understand, unless I conceive three things: 1. the sign; 2. the individual signified; 3. something which determines me to discover the genus or species of the individual of which I must think - in other words, the idea of the genus or species to which the individual signed with the sign belongs.

205. Moreover, with the words, or more generally with the signs which express universals, I do two things:
1. By means of these signs I am led and induced to think of any individual of the given genus and species. For example, with the word man, which indicates for me an individual of the human species, I can mentally concentrate on any particular man, true or imaginary, or I can apply the word man to any particular human being I wish.
This is the first benefit afforded by universals, the first step taken by the spirit as it descends from the species or genus to the individual. But what I have said so far shows perfectly well that I cannot make first use of these names by employing a single idea, that is, the idea of individuals. I need two ideas, that of individuals and that of the species or of the genus to which they belong. Consequently, these ideas of species and genus cannot be mere names. The same is shown by considering the second use we make of universal terms.

2. I also form theories for myself with universal names, that is, I reason in an abstract, universal way without referring to individuals.
In this second use, individuals are either completely excluded and abandoned or act only as signs to assist my spirit in reasoning, but without their constituting in any way the subject about which I reason. Stewart refers to this with an example connected with the use geometricians make of shapes. When a geometrician draws a triangle on the blackboard to demonstrate a universal proposition - for example, the sum of the three internal angles is equal to 180° or two right angles - he uses the individual triangle only as a sign to assist his abstract argument. The proof he gives is not more applicable to the individual triangle than to any other, but it is applicable to all triangles in general. The object of the geometrician's thought is not the specific individual, which is merely a sign, an example, an aid to his thinking. He is thinking of something else, that is, the universal truth he intends to discover and which he does discover with the help of signs, although its nature is totally different from that of signs.

206. Stewart gets close to the truth, yet skirts it, like Horace's charioteer who rounds the half-way mark without touching it. Certainly, if Stewart had hit that half-way mark, his system would have been ruined. He says that individuals have no part in universal reasoning; if they are present, they often do nothing more than hinder and disrupt the flow of the argument. He states all this in the same section of his works in which he discusses universals, but without realising that one single fact of this nature is sufficient to disprove his basic theories. In referring to cases where arbitrary signs are used to support arguments, as in algebra, he says:

He had made the same remark in repeating his opinion about universals:

When an author has committed himself to a false teaching, it is incredible how many contradictions he is obliged to accept, how many mistakes he will pardon himself to ensure that his argument has some sort of appeal. The more intelligent he is, the more his mistakes lead him astray. At this stage, it is worth following his errors carefully and closely examining the track of his tortuous way through the huge labyrinth. There is so much to learn from others' dangers. This is why I take the liberty of pointing out yet again an odd fallacy of reason in a passage of Stewart.

207. Stewart maintains: 'Only individuals can be the object of our thought; what we call general ideas are pure words or signs.' However, in formulating the difficulty for himself, he writes: 'How, in the light of what has been said, are general arguments possible?' To avoid the quandary, he attempts to prove the strange proposition 'that we are able to reason using words, regardless of what the words stand for.'

If his theory is true, such a proposition is, in fact, required. As he did not make use of words expressing individuals in universal reasoning, he had to maintain one of these two statements: either universal words have no meaning or there is something universal, the object of our thought, which these words signify. If the second statement is excluded, the first must be upheld.
To demonstrate this with an example, it was necessary, in my view, to take a universal argument and replace the universal terms which composed it with other random universal terms and see if it still retained some meaning. If the universal terms of an argument are merely valueless signs for us, as Stewart claims, our use of some rather than other signs is totally irrelevant because we pay no heed to their relationships with the things signified.

208. Stewart, however, does not attempt this because it would have been impossible. Instead he offers the following advice about reasoning (I leave it to any common-sense person to say if he is right). He takes some particular reasoning, removes from it the names of the individuals, and replaces these with other names or signs of individuals. He then turns round and says: 'Look, I have changed the names but the reasoning retains its original meaning.' He then infers that we can reason simply by using signs without attributing any value to them. But the only correct conclusion that can be drawn is that in one particular reasoning the names of the individuals may be changed at will, and can also be replaced by common nouns. Stewart's proposition proves nothing more than this.

This is his example.

First, I deny Stewart's assertion that our attention, in the case he proposes, is focused upon simple signs.
I accept that the parties who bring a case before a judge may be designated either by real or fictitious names or by letters of the alphabet. But this merely means that, when dealing with arbitrary names such as proper nouns, we can use whichever we wish. This indifference however is relevant to the signs, not to the ideas. The judge's thinking is directed and focused on the idea signified and does not dwell on the sign itself with which he is not concerned in any way provided that altering the sign does not mean altering the idea and that the first sign is replaced by another which is capable of representing the same idea. This is much easier in the case of proper nouns which have a purely arbitrary connection with the thing designated. Thus, a thing can be designated by any word, by a letter, a syllable, a word of a number of syllables, or by any sign. This is not the case with common nouns or universal terms - or at least the fact does not take place to the same extent. It occurs, however, whenever different synonyms can be found to express the same common notion. This proves that one sign can be substituted for another, but only when ideas do not change as a result. Reasoning is based upon ideas, not upon signs, which have value only to the degree that they signify and suggest ideas. This rules out the possibility of reasoning by simple signs unconnected to any idea. Signs are in fact arbitrary; ideas cannot be arbitrary. Signs may alter provided ideas remain the same. Stewart's example, therefore, proves the exact opposite of what he intended.

209. This will become more obvious the more we study his example. The judge does not need to know the real names of the contending parties, because he does not need to know the individuals themselves in their private relationships outside the case. It is sufficient if he knows them in relationship to the case in hand. The real names allow him to know them as individuals; the fictitious names, or the letters of the alphabet that stand for them, allows him to know them as belonging to a genus of things, that is, as 'persons who have the kind of relations that arise from the case they are bringing, nothing more'. To know them in this second way, the judge must possess universal, abstract ideas; relationships between individuals are merely universal ideas abstracted from real individuality. People who possess some rather than other relationships belong to an accidental species formed from such relationships. Consequently, in replacing the real names of the parties by fictitious names, the only alteration that has taken place in the judge's idea is that an individual idea has been replaced by a generic idea. Stewart was trying to prove with his example that there is no need for ideas of genus, but he has, in fact, given us the best demonstration of the opposite. He intended to show that reasoning can be constructed without the need for general ideas. Instead, he has succeeded in showing us that it is possible to reason without individual ideas, but not without general ideas which may stand alone. This is the outcome of his argument.

Individuals, therefore, are not the only object of human thought nor can signs take the place of universal ideas. The mind, if it is to reason, needs universal much more than individual ideas. It is possible to carry forward an argument which contains no individuals (as in Stewart's earlier example [App., no. 11]) but impossible to conceive how any reasoning can be constructed without universal ideas. Even where reasoning is concerned solely with individuals, they must be considered as endowed with common qualities or common relationships.

210. It is impossible, therefore, despite the application and ingenuity of Scottish philosophy, to eradicate universal ideas which could not even be discussed if they had never existed. This school cannot be praised for eradicating this problem (as it wrongly thought) which Stewart admits as having always been one of the most difficult in metaphysics, that is, the problem of the origin of universal ideas: the problem which, to express it as succinctly as possible, states: 'The human mind cannot form universal ideas for itself without a judgment. But it cannot form a judgment unless it already has universal ideas. It is necessary, therefore, to grant the presence in us of some innate universal idea prior to all our judgments. If nothing innate is acceptable, some other explanation of the difficulty must be found.'
Either way, philosophy is obliged to solve this problem. But the study I have so far undertaken of systems in which an attempt has been made to explain the workings of the spirit without accepting any or almost any(133) innate element in it, shows that they are unable to solve the core issue and that their authors have not even understood it sufficiently.

(115) How poor modern philosophy is to use such language!

(116) Eléments de la Philosophie de l'esprit humain, chap. 4, section 2.

(117) Eléments de la Philosophie de l'esprit humain, chap. 4, section 1.

(118) To reply: 'Isolating mentally is not genuine isolation. The argument, therefore, is false' would show that the question at issue had not been understood. I am referring to intellectual operations of the human spirit, to what occurs in the mind, and not outside it. Relative to the mind, isolating and uniting means conceiving, separately or as a whole, the object which we are thinking of.

(119) Such confusion is, I think, common among modern philosophers. Unable to solve the first question, they introduce the second and unload its absurdities on the first.

(120) Thus, it is the essence of a thing which forms the genus or the species, not the genus or the species which forms the essence. The idea of genus or species presents us with a group - although indeterminate and indefinite - of at least possible individuals. The essence of a thing is utterly simple and unitary.

(121) Strictly speaking, nominal essence should be applied to an essence in which the name alone forms the genus, e.g. 'the genus of those called Peter, Paul, etc.' would be a genus based solely on a thing's name. Comparing this nominal essence with other essences, this genus with other genera, we can easily see how it differs from all other essences and how this genus differs from all other genera. These things cannot therefore be rolled into one as Stewart would wish.

(122) Eléments de la Philosophie de l'esprit humain, chap. 4, section 3.

(123) Ibid.

(124) It is absurd to say that a sensation is altered: it is extremely particular and would first have to be destroyed in order to be altered. Thought, on the other hand, has an object, or idea provided with universal and particular elements. The idea, in so far as it is universal, can be determined and particularised in various ways, and this perhaps could be called 'taking on another form.'

(125) Eléments de la Philosophie de l'esprit humain, chap. 4, section 2.

(126) If we isolate the question of the necessity of language from that of the nature of universals, it is not as difficult as Professor Stewart appears to believe to discover Locke's opinion on this subject. Stewart accuses Locke of employing odd and seldom used expressions in this matter and thus allowing himself to be saddled with contradictory views. This is true, but I do not think that the contradiction lies where Stewart says it does. Stewart finds it contradictory that, in certain passages, Locke does not judge language indispensable to the workings of the intelligence, although he is not a realist. Locke admits that universals are something in the spirit of the thinker, a view which is independent of that relating to the necessity for language. In fact, we may hold that universals are objects of the mind (entia rationis) and at the same time hold, if we wish, both that language is necessary and not necessary for the spirit to form these objects, that is, these ideas which have a nature all of their own.What can rightly be said about Locke is, I think, that he did not get to the root of either question, and that the ridicule heaped on his philosophy by Doria, Martino Scriblero and many subsequent critics is fully justified.

(127) But what if these same attributes are merely a word?

(128) Eléments de la Philosophie de l'esprit humain, chap. 4, section 2.

(129) Does this amount to nothing? Explaining which resemblance is precisely the nub of the question (cf. 180-187).

(130) Eléments de la Philosophie de l'esprit humain, chap. 4, section 2.

(131) Ibid. section 3.

(132) Eléments de la Philisophie de l'esprit humain, chap. 4, section 2.

(133) I say almost any because, in conceding that we come to know bodies, not because sensations offer images of them, but by a kind of inspiration (a faculty sui generis which enables us to perceive bodies when sensations occur), the Scottish school accepts rather more of what is innate than the schools of Locke and Condillac. It admits a new, although obscure and completely mysterious power.