When we come to words of which the logical concept is a complex relation, an obscure or intangible attribute, the defects of the popular conception and its tendencies to change and confusion, are of the greatest practical importance. Take such words as _Monarchy_, _tyranny_, _civil freedom_, _freedom of contract_, _landlord_, _gentleman_, _prig_, _culture_, _education_, _temperance_, _generosity_. Not merely should we find it difficult to give an analytic definition of such words: we might be unable to do so, and yet flatter ourselves that we had a clear understanding of their meaning. But let two men begin to discuss any proposition in which any such word is involved, and it will often be found that they take the word in different senses. If the relation expressed is complex, they have different sides or lines of it in their minds; if the meaning is an obscure quality, they are guided in their application of it by different outward signs.

Monarchy, in its original meaning, is applied to a form of government in which the will of one man is supreme, to make laws or break them, to appoint or dismiss officers of state and justice, to determine peace or war, without control of statute or custom. But supreme power is never thus uncontrolled in reality; and the word has been extended to cover governments in which the power of the titular head is controlled in many different modes and degrees. The existence of a head, with the title of King or Emperor, is the simplest and most salient fact: and wherever this exists, the popular concept of a monarchy is realised. The President of the United States has more real power than the Sovereign of Great Britain; but the one government is called a Republic and the other a Monarchy. People discuss the advantages and disadvantages of monarchy without first deciding whether they take the word in its etymological sense of unlimited power, or its popular sense of titular kingship, or its logical sense of power definitely limited in certain ways. And often in debate, monarchy is really a singular term for the government of Great Britain.

_Culture_, _religious_, _generous_, are names for inward states or qualities: with most individuals some simple outward sign directs the application of the word--it may be manner, or bearing, or routine observances, or even nothing more significant than the cut of the clothes or of the hair. Small things undoubtedly are significant, and we must judge by small things when we have nothing else to go by: but instead of trying to get definite conceptions for our moral epithets, and suspending judgment till we know that the use of the epithet is justified, the trifling superficial sign becomes for us practically the whole meaning of the word. We feel that we must have a judgment of some sort at once: only simple signs are suited to our impatience.

It was with reference to this state of things that Hegel formulated his paradox that the true abstract thinker is the plain man who laughs at philosophy as what he calls abstract and unpractical. He holds decided opinions for or against this or the other abstraction, _freedom_, _tyranny_, _revolution_, _reform_, _socialism_, but what these words mean and within what limits the things signified are desirable or undesirable, he is in too great a hurry to pause and consider.

The disadvantages of this kind of "abstract" thinking are obvious.

The accumulated wisdom of mankind is stored in language. Until we have cleared our conceptions, and penetrated to the full meaning of words, that wisdom is a sealed book to us. Wise maxims are interpreted by us hastily in accordance with our own narrow conceptions. All the vocables of a language may be more or less familiar to us, and yet we may not have learnt it as an instrument of thought. Outside the very limited range of names for what we see and use in the daily routine of life, food and clothes and the common occupations of men, words have little meaning for us, and are the vehicles merely of thin preconceptions and raw prejudices.

The remedy for "abstract" thinking is more thinking, and in pursuing this two aims may be specified for the sake of clearness, though they are closely allied, and progress towards both may often be made by one and the same operation. (1) We want to reach a clear and full conception of the meaning of names as used now or at a given time.

Let us call this the _Verification of the Meaning_. (2) We want to fix such conceptions, and if necessary readjust their boundaries. This is the province of _Definition_, which cannot be effectually performed without _Scientific Classification_ or _Division_.

I.--VERIFICATION OF THE MEANING--DIALECTIC.

This can only be done by assembling the objects to which the words are applied, and considering what they have in common. To ascertain the actual connotation we must run over the actual denotation. And since in such an operation two or more minds are better than one, discussion or dialectic is both more fruitful and more stimulating than solitary reflection or reading.

The first to practise this process on a memorable scale, and with a distinct method and purpose, was Socrates. To insist upon the necessity of clear conceptions, and to assist by his dialectic procedure in forming them, was his contribution to philosophy.

His plan was to take a common name, profess ignorance of its meaning, and ask his interlocutor whether he would apply it in such and such an instance, producing one after another. According to Xenophon's _Memorabilia_ he habitually chose the commonest names, _good_, _unjust_, _fitting_, and so forth, and tried to set men thinking about them, and helped them by his questions to form an intelligent conception of the meaning.

For example, what is the meaning of injustice? Would you say that the man who cheats or deceives is unjust? Suppose a man deceives his enemies, is there any injustice in that? Can the definition be that a man who deceives his friends is unjust? But there are cases where friends are deceived for their own good: are these cases of injustice?

A general may inspirit his soldiers by a falsehood. A man may cajole a weapon out of his friend's hand when he sees him about to commit suicide. A father may deceive his son into taking medicine. Would you call these men unjust? By some such process of interrogation we are brought to the definition that a man is unjust who deceives his friends to their hurt.

Observe that in much of his dialectic the aim of Socrates was merely to bring out the meaning lying vague and latent, as it were, in the common mind. His object was simply what we have called the verification of the meaning. And a dialectic that confines itself to the consideration of what is ordinarily meant as distinct from what ought to be meant may often serve a useful purpose. Disputes about words are not always as idle as is sometimes supposed. Mr. H. Sidgwick truly remarks (_a propos_ of the terms of Political Economy) that there is often more profit in seeking a definition than in finding it.

Conceptions are not merely cleared but deepened by the process. Mr.

Sidgwick's remarks are so happy that I must take leave to quote them: they apply not merely to the verification of ordinary meaning but also to the study of special uses by authorities, and the reasons for those special uses.

"The truth is--as most readers of Plato know, only it is a truth difficult to retain and apply--that what we gain by discussing a definition is often but slightly represented in the superior fitness of the formula that we ultimately adopt; it consists chiefly in the greater clearness and fulness in which the characteristics of the matter to which the formula refers have been brought before the mind in the process of seeking for it. While we are apparently aiming at definitions of terms, our attention should be really fixed on distinctions and relations of fact. These latter are what we are concerned to know, contemplate, and as far as possible arrange and systematise; and in subjects where we cannot present them to the mind in ordinary fulness by the exercise of the organs of sense, there is no way of surveying them so convenient as that of reflecting on our use of common terms.... In comparing different definitions our aim should be far less to decide which we ought to adopt, than to apprehend and duly consider the grounds on which each has commended itself to reflective minds. We shall generally find that each writer has noted some relation, some resemblance or difference, which others have overlooked; and we shall gain in completeness, and often in precision, of view by following him in his observations, whether or not we follow him in his conclusions."[2]

Mr. Sidgwick's own discussions of _Wealth_, _Value_, and _Money_ are models. A clue is often found to the meaning in examining startlingly discrepant statements connected with the same leading word. Thus we find some authorities declaring that "style" cannot be taught or learnt, while others declare that it can. But on trying to ascertain what they mean by "style," we find that those who say it cannot be taught mean either a certain marked individual character or manner of writing--as in Buffon's saying, _Le style c'est l'homme meme_--or a certain felicity and dignity of expression, while those who say style can be taught mean lucid method in the structure of sentences or in the arrangement of a discourse. Again in discussions on the rank of poets, we find different conceptions of what constitutes greatness in poetry lying at the root of the inclusion of this or the other poet among great poets. We find one poet excluded from the first rank of greatness because his poetry was not serious; another because his poetry was not widely popular; another because he wrote comparatively little; another because he wrote only songs or odes and never attempted drama or epic. These various opinions point to different conceptions of what constitutes greatness in poets, different connotations of "great poet". Comparing different opinions concerning "education" we may be led to ask whether it means more than instruction in the details of certain subjects, whether it does not also import the formation of a disposition to learn or an interest in learning or instruction in a certain method of learning.

Historically, dialectic turning on the use of words preceded the attempt to formulate principles of Definition, and attempts at precise definition led to Division and Classification, that is to systematic arrangement of the objects to be defined. Attempt to define any such word as "education," and you gradually become sensible of the needs in respect of method that forced themselves upon mankind in the history of thought. You soon become aware that you cannot define it by itself alone; that you are beset by a swarm of more or less synonymous words, _instruction_, _discipline_, _culture_, _training_, and so on; that these various words represent distinctions and relations among things more or less allied; and that, if each must be fixed to a definite meaning, this must be done with reference to one another and to the whole department of things that they cover.

The first memorable attempts at scientific arrangement were Aristotle's treatises on Ethics and Politics, which had been the subjects of active dialectic for at least a century before. That these the most difficult of all departments to subject to scientific treatment should have been the first chosen was due simply to their preponderating interest: "The proper study of mankind is man". The systems of what are known as the Natural Sciences are of modern origin: the first, that of Botany, dates from Cesalpinus in the sixteenth century. But the principles on which Aristotle proceeded in dividing and defining, principles which have gradually themselves been more precisely formulated, are principles applicable to all systematic arrangements for purposes of orderly study. I give them in the precise formulae which they have gradually assumed in the tradition of Logic.

The principles of Division are often given in Formal Logic, and the principles of Classification in Inductive Logic, but there is no valid reason for the separation. The classification of objects in the Natural Sciences, of animals, plants, and stones, with a view to the thorough study of them in form, structure, and function, is more complex than classifications for more limited purposes, and the tendency is to restrict the word classification to these elaborate systems. But really they are only a series of divisions and subdivisions, and the same principles apply to each of the subordinate divisions as well as to the division of the whole department of study.

II.--PRINCIPLES OF DIVISION OR CLASSIFICATION AND DEFINITION.

Confusion in the boundaries of names arises from confused ideas regarding the resemblances and differences of things. As a protective against this confusion, things must be clearly distinguished in their points of likeness and difference, and this leads to their arrangement in systems, that is, to division and classification. A name is not secure against variation until it has a distinct place in such a system as a symbol for clearly distinguished attributes. Nor must we forget, further, that systems have their day, that the best system attainable is only temporary, and may have to be recast to correspond with changes of things and of man's way of looking at them.

The leading principles of DIVISION may be stated as follows:--

I. Every division is made on the ground of differences in some attribute common to all the members of the whole to be divided.

This is merely a way of stating what a logical division is. It is a division of a generic whole or _genus_, an indefinite number of objects thought of together as possessing some common character or attribute. All have this attribute, which is technically called the _fundamentum divisionis_, or generic attribute. But the whole is divisible into smaller groups (_species_), each of which possesses the common character with a difference (_differentia_). Thus, mankind may be divided into White men, Black men, Yellow men, on ground of the differences in the colour of their skins: all have skins of some colour: this is the _fundamentum divisionis_: but each subdivision or species has a different colour: this is the _differentia_. Rectilineal figures are divided into triangles, quadrangles, pentagons, etc., on the ground of differences in the number of angles.

Unless there is a _fund. div._, _i.e._, unless the differences are differences in a common character, the division is not a logical division. To divide men into Europeans, opticians, tailors, blondes, brunettes, and dyspeptics is not to make a logical division. This is seen more clearly in connexion with the second condition of a perfect division.

II. In a perfect division, the subdivisions or species are mutually exclusive.

Every object possessing the common character should be in one or other of the groups, and none should be in more than one.

Confusion between classes, or overlapping, may arise from two causes. It may be due (1) to faulty division, to want of unity in the _fundamentum divisionis_; (2) to the indistinct character of the objects to be defined.

(1) Unless the division is based upon a single ground, unless each species is based upon some mode of the generic character, confusion is almost certain to arise. Suppose the field to be divided, the objects to be classified, are three-sided rectilineal plane figures, each group must be based upon some modification of the three sides. Divide them into equilateral, isosceles, and scalene according as the three sides are all of equal length, or two of equal length, or each of different length, and you have a perfect division. Similarly you can divide them perfectly according to the character of the angles into acute-angled, right-angled and obtuse-angled. But if you do not keep to a single basis, as in dividing them into equilateral, isosceles, scalene, and right-angled, you have a cross-division. The same triangle might be both right-angled and isosceles.

(2) Overlapping, however, may be unavoidable in practice owing to the nature of the objects. There may be objects in which the dividing characters are not distinctly marked, objects that possess the differentiae of more than one group in a greater or less degree. Things are not always marked off from one another by hard and fast lines.

They shade into one another by imperceptible gradations. A clear separation of them may be impossible. In that case you must allow a certain indeterminate margin between your classes, and sometimes it may be necessary to put an object into more than one class.

To insist that there is no essential difference unless a clear demarcation can be made is a fallacy. A sophistical trick called the _Sorites_ or Heap from the classical example of it was based upon this difficulty of drawing sharp lines of definition. Assuming that it is possible to say how many stones constitute a heap, you begin by asking whether three stones form a heap. If your respondent says No, you ask whether four stones form a heap, then five, and so on and he is puzzled to say when the addition of a single stone makes that a heap which was not a heap before. Or you may begin by asking whether twenty stones form a heap, then nineteen, then eighteen, and so on, the difficulty being to say when what was a heap ceases to be so.

Where the objects classified are mixed states or affections, the products of interacting factors, or differently interlaced or interfused growths from common roots, as in the case of virtues, or emotions, or literary qualities, sharp demarcations are impossible.

To distinguish between wit and humour, or humour and pathos, or pathos and sublimity is difficult because the same composition may partake of more than one character. The specific characters cannot be made rigidly exclusive one of another.

Even in the natural sciences, where the individuals are concrete objects of perception, it may be difficult to decide in which of two opposed groups an object should be included. Sydney Smith has commemorated the perplexities of Naturalists over the newly discovered animals and plants of Botany Bay, in especial with the _Ornithorynchus_,--"a quadruped as big as a large cat, with the eyes, colour, and skin of a mole, and the bill and web-feet of a duck--puzzling Dr. Shaw, and rendering the latter half of his life miserable, from his utter inability to determine whether it was a bird or a beast".

III. The classes in any scheme of division should be of co-ordinate rank.

The classes may be mutually exclusive, and yet the division imperfect, owing to their not being of equal rank. Thus in the ordinary division of the Parts of Speech, parts, that is, of a sentence, Prepositions and Conjunctions are not co-ordinate in respect of function, which is the basis of the division, with Nouns, Adjectives, Verbs, and Adverbs.

The preposition is a part of a phrase which serves the same function as an adjective, _e.g._, _royal_ army, army _of the king_; it is thus functionally part of a part, or a particle. So with the conjunction: it also is a part of a part, _i.e._, part of a clause serving the function of adjective or adverb.

IV. The basis of division (_fundamentum divisionis_) should be an attribute admitting of important differences.

The importance of the attribute chosen as basis may vary with the purpose of the division. An attribute that is of no importance in one division, may be important enough to be the basis of another division.

Thus in a division of houses according to their architectural attributes, the number of windows or the rent is of little importance; but if houses are taxed or rated according to the number of windows or the rent, these attributes become important enough to be a basis of division for purposes of taxation or rating. They then admit of important differences.

That the importance is relative to the purpose of the division should be borne in mind because there is a tendency to regard attributes that are of importance in any familiar or pre-eminent division as if they had an absolute importance. In short, disregard of this relativity is a fallacy to be guarded against.

In the sciences, the purpose being the attainment and preservation of knowledge, the objects of study are divided so as to serve that purpose. Groups must be formed so as to bring together the objects that have most in common. The question, Who are to be placed together?

in any arrangement for purposes of study, receives the same answer for individuals and for classes that have to be grouped into higher classes, namely, Those that have most in common. This is what Dr. Bain happily calls "the golden rule" of scientific classification: "Of the various groupings of resembling things, preference is given to such as have the greatest number of attributes in common". I slightly modify Dr. Bain's statement: he says "the most numerous and the most important attributes in common". But for scientific purposes number of attributes constitutes importance, as is well recognised by Dr. Fowler when he says that the test of importance in an attribute proposed as a basis of classification is the number of other attributes of which it is an index or invariable accompaniment. Thus in Zoology the squirrel, the rat, and the beaver are classed together as Rodents, the difference between their teeth and the teeth of other Mammalia being the basis of division, because the difference in teeth is accompanied by differences in many other properties. So the hedge-hog, the shrew-mouse, and the mole, though very unlike in outward appearance and habits, are classed together as Insectivora, the difference in what they feed on being accompanied by a number of other differences.

_The Principles of_ DEFINITION. The word "definition" as used in Logic shows the usual tendency of words to wander from a strict meaning and become ambiguous. Throughout most of its uses it retains this much of a common signification, the fixing or determining of the boundaries of a class[3] by making clear its constituent attributes. Now in this making clear two processes may be distinguished, a material process and a verbal process. We have (1) the clearing up of the common attributes by a careful examination of the objects included in the class: and we have (2) the statement of these common attributes in language. The rules of definition given by Dr. Bain, who devotes a separate Book in his Logic to the subject of Definition, concern the first of these processes: the rules more commonly given concern mainly the second.

One eminent merit in Dr. Bain's treatment is that it recognises the close connexion between Definition and Classification. His cardinal rules are reduced to two.

I. _Assemble for comparison representative individuals of the class._

II. _Assemble for comparison representative individuals of the contrasted class or classes._

Seeing that the contrasted classes are contrasted on some basis of division, this is in effect to recognise that you cannot clearly define any class except in a scheme of classification. You must have a wide _genus_ with its _fundamentum divisionis_, and, within this, _species_ distinguished by their several _differentiae_.

Next, as to the verbal process, rules are commonly laid down mostly of a trifling and obvious character. That "a definition should state neither more nor less than the common attributes of the class," or than the attributes signified by the class-name, is sometimes given as a rule of definition. This is really an explanation of what a definition is, a definition of a definition. And as far as mere statement goes it is not strictly accurate, for when the attributes of a genus are known it is not necessary to give all the attributes of the species, which include the generic attributes as well, but it is sufficient to give the generic name and the differentia. Thus Poetry may be defined as "a Fine Art having metrical language as its instrument". This is technically known as definition _per genus et differentiam_. This mode of statement is a recognition of the connexion between Definition and Division.