BOUGHT WITH THE INCOME FROM THE SAGE ENDOWMENT FUND THE GIFT OF iieniril W. Sage 1S9Z A.S.ZUA i^^^kL *-HS STREET 1892 All rights reserved PREFACE The object of this book is not complacently to put forward a recondite system of certainties, but to raise and discuss from an ordinary point of view some questions which seem to me interesting, many-sided, and hard to answer. The faith in which it is written is that, so far as people gene- rally dislike philosophy, it is rather the name than the thing itself that they dislike — the name, with its supposed pretensions of superiority. From modesty alone many people hide from themselves their own philosophical inclinations, and start with a prejudice against any writer who confesses such inclinations openly. Perhaps no complete defence is possible. However, the truth remains (and the charitable reader will see it) that I do not think there is at present room for anything like authority in treating these questions. They present to me, therefore, not a chance of playing the oracle, but at most a chance of joining others in their own pursuit of truth. PREFACE Some charitable consideration for the style may also be claimed in other respects. For instance, technical terms cannot entirely be dispensed with, try as hard as one may to write the plainest English. And, again, satisfactory examples are difficult to select, for, the clearer and more familiar they are, the more they tend to be trivial. In these and similar matters a writer has to find his way as well as he can between opposite dangers, and can- not expsct entire success in doing so. I am greatly indebted to a critic {Mons. G. Fonsegrive, in the Revue Philosophique) who pointed out some defects in a book I published some years ago.' It was in the attempt to remedy those defects that much of the view here taken was reached. And I am still more grateful to Mr. Carveth Read for the important corrections he has sug- gested, both in the proof-sheets and • in earlier stages of the work. April, 1892. ' Fallacies (Intemational Scientific Series). CONTENTS TAGS The meaning of 'Ambiguity.'' ' Unreal Distinctness ' as its cause. CHAP. I. PRELIMINARY SKETCH OF THE SUBJECT. I ,, II. THE NATURE OF ROUGH DISTINCTION . I4 The extent of Unreal Distinctness that exists in Language and Thought. CHAP. III. PHILOSOPHY AND COMMON-SENSE . . 30 ,, IV. RIVAL IDEALS 4I „ V. UNREALITY . . ... 58 ,, VI. IS NATURE CONTINUOUS THROUGHOUT?. 7 1 ,, VII. COMMON-SENSE AS DIVIDED AGAINST ITSELF ... .80 ,, VIII. SPOILT WOEDS .... tJ8 The harm that is caused by Unreal Distinctness. CHAP. IX. THE EFFECTS OF AMBIGUITY . . 98 ,, X. THE DEMAND FOR STRICT DEFINITION. I IJ Remedy: the escape jrom the demand for strict dejfinition by means of the notion of ' sufficient ' definiteness, or the irrelevance of the demand. CHAP. XL THE 'SPECIAL OCCASION' . . .131 „ XII. THE RELATIVITY OF DISTINCTIONS . 143 viii CONTENTS PAGE CHAP. Xm. SOME ILLUSTRATIONS ISO ,, XIV. 'PROPER' AND 'GENERAL' NAMES . . . l68 „ XV. 'REFERENCE-NAMES' ... -179 ,, XVI. THE CRITICISM OF DISTINCTIONS IN RELATION TO BELIEF ... 191 ,, XVII. SCEPTICISM AND CONCILIATION . . . 202 ,, XVIIL GENERAL SUMMARY . . . 22$ APPENDIX Nameable Things . . ... 253 Thought and Language . . . . 256 Thinghood .... . .257 The Continuity of Nature . . 259 MilPs 'Inductive Canons ' . . . .261 Some Examples from Philosophy 264 Middle Term and Major Premiss . . . . 267 INDEX . . 271 DISTINCTION CHAPTER I PRELIMINARY SKETCH OF THE SUBJECT As everyone knows, there are plenty of topics which are full of a sort of amateur-philosophical interest ; topics which, on the one hand, are marked off from mere gossip, or anecdote, or talk about places, as being more general and less matter-of- fact than these, and, on the other hand, are marked off from technical discussions as being freely carried on in our everyday language, and, on the surface at least, requiring no special education or training. In the hurry and press of business, and of the wider business of living our lives successfully, our attention is mostly fastened on matters of what is called practical interest and importance — on the things that we are somehow bound to know under B DISTINCTION penalty of failing, or at least of living rather restricted and unsatisfactory lives. Few are the people, however, who care to draw the line very sharply between knowledge of this sort and know- ledge that lacks practical value ; we trust rather to our common-sense tact of the moment to keep us away from barren fields of speculation ; and all except the dullest and most unfortunate of us are well aware that plenty of subjects that lie a little aside from our regular habits of pursuit would be found interesting and important if we had time to pursue them. So it comes about that we play with some of these less pressing subjects now and again, unbend our minds over them, or even take sides and discuss them with real though fleeting interest. Such subjects come up in the course of conversation, or we light upon them by accident in a review or a book ; and every time this happens we bestir ourselves afresh, and resolve that now at last we will make the matter finally clear to our own comprehension ; we will make sure once for all, let us say, what Darwin's theory was, and how much of it survives in spite of criticism ; or what is exactly the weakness of Fair-trade proposals, or of existing Socialist theories ; we will come to terms with ourselves in regard to some puzzling item in our professed political or religious faith. PREUMINARY SKETCH And yet, somehow, next time the question assails us, after a lapse of weeks or months, we are rather vexed to find that our good resolution came to very little after all ; we have forgotten the clue we thought we had found, and the old puzzle recurs, and is just as fresh and just as elusive as ever. In dealing with all these recurring puzzles — these numerous wordy or 'notional' topics of dispute — the attempt to unravel ambiguities of language has long been known to be useful. It is the first vague general clue we discover, and it remains our best safeguard when the last word is said ; and though to acquire the fullest and fairest use of this clue is a matter of endless education, yet a good deal can certainly be done by setting ourselves upon the right track for acquiring it, from the beginning. That, accordingly, is the main purpose of the present book : we are to see as well as we can what this aim at clearness of language means and involves, and we are to review and revise some of our present ideas as to attaining it. One of the first things, therefore, to ask our- selves is, what we shall mean by ambiguity, and here I have a suggestion to make that I hope will be found of service. Roughly, an 'ambiguous' word may be defined as a word with two or more B 2 DISTINCTION meanings ; it is not, however, the bare fact that a word has two or more meanings that makes it ambiguous in any effectual sense, but the fact that its two or more meanings are in practice confused- The examples of ambiguous (or ' equivocal ') words that are commonly given in books on logic rather draw attention away from the most effectual kind of ambiguity. Names like pound, or foot, or post, may certainly be said to have two or more mean- ings, but a word is not ambiguous as used in assertion, unless there be real doubt which sense is intended. And such doubt hardly ever arises in the case of words that are well known to have several different senses, for then the need of letting the context explain them is evident to assertor and audience equally. You can make puns upon a word VCks pound, or foot, ox post, but you can hardly by means of them mislead either yourself or the most uncritical audience. Instead, therefore, of thinking of ambiguity as if it lay in the mere existence of two distinct meanings for the same word, it is better to view it as consisting in vague- ness of outline, which is chiefly shown in uncertainty as to the cases supposed to come under the name. In one of its meanings, for instance, a word like civilised might include the founders of Carson ' City,' or the miners of Euchre Flat ; used in a PRELIMINARY SKETCH somewhat narrower sense it might exclude them ; or, at any rate, doubtful examples of its applica- bility are easy to find — cases, that is, where different people would use the word with different meanings. Or think of the ambiguities that arise out of words like hero, gentleman, artist, patriot, and, in fact, all our commonest epithets implying praise, or blame, or contempt. Different people put different meanings upon them, and so they become misleading. That the absence of clear definition is the source of all the most effective ambiguity, will perhaps become plainer as we proceed to notice examples of various sorts. But a reason may also be given which will help us to understand the truth more clearly. Ambiguity, like every insidious fault, is most effective where it is least suspected, least easy to see at a careless glance. And natur- ally this occurs, to the greatest extent, not where a word means widely and strikingly different things, like pound and pound, but where the things it means are nearly the same on the surface and only differ in deeper or more occasional ways. Every class-name ' groups together things (or people or ' I here use this to include every general (or descriptive) name. The distinction between the substantive and the adjective or verb is irrelevant. See also pp. 173-5. DISTINCTION cases) which resemble each other in some respects and differ in others ; and the question ■ is always liable to arise whether on some given occasion of its use the resemblance or the difference is the more important. And hence the most effective kind of ambiguity occurs where a word in much of its everyday use is plain and unmistakable, and only becomes ambiguous on comparatively rare occasions. A name like work, a distinction like that between work and play, will serve roughly to illustrate the kind of cases here referred to. For children, for boys at school, for day-labourers, for officials generally, and perhaps for most of those who go six days a week to the place where their money is made, there is seldom any practical doubt about the distinction. The schoolmaster or the employer of labour knows pretty well which of his boys, or which of his hands, are the hardest workers ; the world as a rule finds out its thoroughly idle men. But with some professional and semi- professional classes — poets, and artists generally, for example — the case is different; the best of thieir work is not always that part over which most effort or patience is expended, and though even the greatest genius probably has to undergo some drudgery, there is a point, easily reached as a rule in all artistic work, where drudgery begins to spoil PRELIMINARY SKETCH the results — to make them, as the saying is, smell of the midnight oil. Besides, the artist will often be liable to have forced upon him, from within if not by other people, doubts as to whether even his greatest efforts are properly ' work ' at all, or whether his results at their best have any serious value ; he might have been helping to build lighthouses, or to conquer kingdoms, instead of staying at home ' to play with paper like a child.' But a distinction like that between work and play is a comparatively obvious, and therefore comparatively harmless, source of confusion. The finer degrees of the diffi- culty may best be illustrated by some case where doubts as to the line never actually arise in prac- tice, but only in theorising — for example, our fundamental notions of physics, where distinctions that are perfectly valid for all practical purposes are carried over into regions where their validity holds no longer. The distinction between motion and rest is one of the simplest and most familiar of these cases ; absolute rest is unknown to us, yet all motion is measured by rest. And it is notorious that many of the permanent troubles of metaphysics arise from just this kind of ambiguity. That, therefore, is the reason why distinction, and especially rough distinction, is the central sub- ject of this book. Ambiguity, in its most effective DISTINCTION and troublesome form, arises out of the ' real ' ' roughness of distinctions that are drawn by lan- guage as if they were perfectly sharp. We are, therefore, to review these cases both in a general way and in some detail, hoping by this process to systematise what we already vaguely know about effective ambiguity, and about the mistakes and puzzles into which it is always leading us. We shall have occasion also to find that some of the views that are apparently held by common-sense require a certain amount of correction or limita- tion ; which is only natural, since common-sense embodies itself in no single or definite creed, but embraces many different stages of insight, none of which, I assume, are wholly beyond the reach of error. It will possibly be a convenience to the reader if I give beforehand ^ a short general sketch of the course of our argument, and mention some of its incidental aims. First, then, an attempt is made to discover the part that is actually played by ambiguity (or rough distinction) in confusing our judgment This is not quite a simple matter to settle. We can notice, indeed, the extent of the use of rough distinctions, and we can discover ' See p. 22. ^ Another and fuller sketch is given in Chap, xviii. p. 225. PRELIMINARY SKETCH pretty clearly the manner in which ambiguity works when it disturbs our judgment at all, but how far our judgment is actually disturbed in given cases it is not so easy to see. In regard to this latter question, I wish at least to recognise the fact that common-sense is not always deceived by indefinite names, but that it often uses a faulty distinction with full or sufficient knowledge of its faults. And, secondly, in the process of getting to understand exactly the error that rough distinction creates, we shall find it necessary to discuss the excuses that may fairly be made sometimes for vagueness. For the use and abuse of rough dis- tinction lie close together, and a knowledge of each will help to explain the other. This enquiry is one that may lead us very far into philosophy, but the interest in it begins long before what is commonly called philosophy is reached. For at every level of thought we are soon brought up against the difficulties that arise out of the attempt to define our words — that is to say, the attempt to draw sharp distinctions where the things distin- guished shade off into one another. These diffi- culties are familiar to everyone, and I hope no apology is needed for trying any new methods we can find for their solution. DISTINCTION Our main purpose, already mentioned, includes an attempt to find a more philosophical method of dealing with rough distinctions, in place of the happy-go-lucky tact that everyone uses, more or less, by the light of Nature ; and of incidental questions that arise and suggest lines of further enquiry there are a considerable number. For instance, we can hardly avoid some general reflections upon the nature of controversy. Not only does controversial matter afford the clearest examples of the error that ambiguity causes, but the criticism of distinctions and definitions is very closely allied to criticism of the soundness of a judgment ; and, conversely, th§ ways of escaping from such criticism (or of justifying vagueness) are also ways of defending an assertion against a disputer. Nearly ' all criticism — nearly all objec- tions to any belief — may without much difficulty be viewed as complaining that some distinction is rougher than the believer fancies ; and to discuss our central question is thus at the same time to ask what is the full force, and what the extent of the weakness, of the objection that a given distinc-' tion is rough. And another subject that is interwoven with ' It is difficult, but not impossible, to see all criticism of judg- ment as criticism of distinction. Seep. 193. PRELIMINARY SKETCH our main enquiry is the everlasting struggle that language carries on against difficulties of expres- sion. We cannot go far into the problems of definition without being forced to see that lan- guage, like many other inventions or results, naturally grows and develops under the constant pressure of various needs — needs that are partly conflicting. It exists by yielding to the living and shifting force of partly opposite aims. Like a machine or an organism, its present state repre- sents a long succession of compromises, of gradu- ally wiser attempts to find the best combination of antagonistic qualities. It must, for example, be ready, and yet not too rough ; its distinctions must be many enough to cope with real differences, and yet few enough for the average man to remember ; it must take its stand upon existing knowledge, and yet be elastic enough to welcome a conquering change of theory ; and any statement must be long enough, and also short enough, to produce the re- quired effect. Hence the faults of language,- like some of the faults of human beings, may be viewed as only virtues exercised upon the wrong occasion. This does not, however, prevent their calling for a remedy, and much of the remedy lies in knowledge — in the fullest answers we can give to the question. DISTINCTION What are the uses and the abuses of rough dis- tinction ? We shall also make a closer acquaintance than mere haphazard experience gives us with the way in which language acts as a drag upon the progress of knowledge. There is a certain over-conservative tendency in our thought which keeps us more under slavery to ' words than we need be. The weight of this incubus has, indeed, been greatly les- sened in the last few centuries, but we have not yet arrived at a wide understanding of Hobbes's saying — that ' words are the counters of wise men and the money of fools.' In discussing the pro- blems of distinction we shall see, better perhaps than in any other way, the truth that words are essentially instruments of expression, or mean just what they are meant to mean and no more. And, lastly, the question of questions in philo- sophy, the question at what point doubt must come to an end, is altered suggestively for us if we admit the truth of the views here taken. For the essence of scepticism is casuistry, or the enquiry after a definition which shall be applicable to actual cases, instead of merely general and abstract. We shall see that such an enquiry can never be perfectly satisfied, and that therein lies the strength of the sceptical attack ; but we shall also see that PRELIMINARY SKETCH 13 it may be irrelevant to some special or passing purpose, and that by such irrelevance the destruc- tive force of scepticism is limited. So short a statement, however, can be of little service, except as a memorandum after the subject has been thoroughly discussed. Its total meaning lies in the details of meaning it covers, and one of the aims of the present book is to give a sufficient number of these details to satisfy the reader that some useful work remains to be done by himself in carrying further and testing more fully with concrete examples the chief suggestions here put forward. The next chapter may be regarded as a neces- sary evil. It attempts to explain shortly the sense in which certain words and phrases are used. For readers familiar with logic, the more important of the explanations are : unreal distinctness (p. 22 and note), applicability (p. 22), and ideal and actual (P- 27). CHAPTER II THE NATURE OF ROUGH DISTINCTION § I. Distinction in general. — By a distinction will here be meant what is commonly meant by that name — a recognition in thought and language of a difference between one 'nameable thing' and another or others ; a difference between A and B, or A and non-A.' The process of distinction is a process of mentally separating this and that object of thought, or any one from all others. To dis- tinguish is to recognise otherness anywhere. No ' The expression ' nameable thing ' is here used, in preference to 'thing,' in order to avoid the associations of the latter word with reality. Whatever we choose to mean by reality, distinctions may be drawn between unrealities just as well as between realities, al- ways provided the former are ' nameable ' and so correspond to what are technically called ' general notions ' — which may be either substantival, adjectival, or verbal. Thus A, non-A, B, &c. , repre- sent general (or descriptive) names — i.e. broadly speaking (see Chap. XV. ), any common nouns, adjectives, and verbs — and if the reader happens to find the practice of using letters troublesome, he can always substitute names for them ; for instance, life and death, alive and dead, live and die. And see Appendix, pp. 253-9. THE NATURE OF ROUGH DISTINCTION 15 matter how large or how small the difference is supposed to be, any recognised difference, as such, will form a distinction ; the separation of two units, for example, is just as much a distinction as that between the most diverse possible things. Dis- tinction of some sort exists, for us, wherever we see plurality, since plurality is a form of ' other- ness.' From this it follows that every descriptive name, as such, implies a distinction. Whatever descriptive meaning it has is due to its separation from a background' — i.e. from all 'other' things (or qualities, or actions). To distinguish is, therefore, so far as it goes, to define. Where any two notions are contrasted we talk of the distinction between them ; where any one notion is contrasted with all others, we talk of its definition ; so that the process of definition includes the process of distinction, and only differs from it in being more sweeping in its extent. To understand distinction thoroughly is at the same time to understand definition. § 2. Rough distinction. — By a ' rough ' distinc- tion we shall mean, as is usually meant, a distinction where the contrasted notions, even at their sharpest (A and non-A), cannot be applied with perfect ' See also p. 100. i6 DISTINCTION exactness to actual cases ; where the actual cases cannot always be classed with strict right as either the one or the other, but where a certain proportion of them belong to a doubtful borderland. The expressions 'difference of degree' and ' difference of kind ' are very commonly used ; and a rough distinction is a distinction that depends on a difference of degree. Words like large, or cold, or heavy are familiar examples ; differences of size, of temperature, of weight, &c., are easily seen to be gradual differences ; whatever grades we recog- nise are plainly artificial — made for convenience. For even admitting that the extreme ends of a given scale are clearly distinct, where do the ' ends ' begin t How are the ends marked off from the intermediate region ? The recognition of a middle portion at all is only our way of confessing a difficulty in separating the two ends sharply from each other. It does for the moment save us, but the difficulty recurs when we ask what are the exact limits of the intermediate region. Between the opposites good and bad, for instance, we insert a vague intermediate region called indifferent; where does this end, or what are the exact limits of middle-age, or of the middle-classes ? § 3. ' Artificially sharp ' distinction. — A distinc- tion drawn too sharply, or made artificially sharp, THE NATURE OF ROUGH DISTINCTION 17 is the same as a rough distinction. Roughness means, in this connection, the opposite of sharpness, and at the first glance it might seem difficult to reconcile these two notions ; but the difficulty, if felt at all, is a merely verbal one, the contradiction depending upon the opposite points of view from which any given distinction may be regarded. A distinction is drawn too sharply where the things or cases distinguished are in fact only roughly distinct — that is to say, where the ' real ' difference is one of degree ; or, on the other hand, a distinction is ' really ' rough when it is artificially sharpened beyond what the facts will justify. When we call the distinction ' too sharp,' we refer to the distin- guishing idea ; when we call the distinction ' rough,' we refer to the facts distinguished. § 4. Recognition of roughness or artificiality. — A further difficulty, however, in connecting rough- ness with artificial sharpness, arises from the fact that the artificiality of a sharp distinction is often so plainly recognised by common-sense that the distinction is not in practice taken sharply at all. Language is full of words and distinctions that are only not considered definite, or distinctions of kind, because their artificiality is admitted by all. Every- one knows that no real line can be strictly drawn be- tween, for instance, a bud and 2i flower, a house and c DISTINCTION a cottage, the rich and the poor man, heat and cold^ and so on ; everyone sees that to make the twenty- first birthday the turning-point of legal status is an artificial proceeding ; many people know that all our epithets distributing praise or blame are ex- ceedingly vague and ' wordy ' ; some people acquire the habit of seeing that any distinction, as such, lies open to sceptical doubts. And common-sense certainly claims to exercise a kind of tact in using openly faulty distinctions ; claims to possess the power of taking and using them lightly, refusing to press for anything like exactness in drawing the line. And then the question arises : Is there, properly speaking, any actual sharpness of distinc- tion at all where the artificiality of the sharpness is clearly seen and discounted ? Is not roughness of distinction something different from artificial sharpness, the latter only occurring where a rough distinction is mistakenly supposed to be sharp } Here, of course, we may take our choice which meaning the term ' artificial sharpness ' shall bear. Let us, then, choose to make it exactly synonymous with roughness of distinction, thus regarding it as a quality of the distinction itself rather than of the way the distinction is taken. This does not dispose of all the difficulty which the exercise of common-sense tact introduces, but it enables us to THE NATURE OF ROUGH DISTINCTION 19 postpone that difficulty to a later place in this book.' § 5. How are rough distinctions possible ? — An- other verbal puzzle may also deserve to be shortly dismissed at the outset. Though everyone knows that rough distinctions do, as a matter of fact, exist, it is not very difficult to raise an apparently logical doubt as to whether their existence is possible. And it seems worth while briefly to meet this objection, if for no other purpose than that of beginning at once to rid ourselves of the wrong kind of fear of self-contradiction. It must be admitted, I think, that the phrase, ' a rough distinction,' is in strictness self-contradictory. Between distinctness and indistinctness there is, in idea, no middle ground. A ' rough ' line is not properly a line at all ; and so far as a distinction is drawn the distinguished things are distinct. But the explanation of the phrase is simple enough, and reminds us at once of numerous other occa- sions where a self-contradictory term is found in practice useful. Their use always is to name an intermediate state, whether a passing stage of development or not. Thus we may speak of ' unconscious hypocrisy,' ' Tory democracy,' or the even more easily understood phenomenon, ' melting ' See especially pp. 83, 135, 147. DISTINCTION snow.' And just as any appreciable quantity of melting snow is partly snow and partly water, so we may steady the phrase, ' a rough distinction,' by remembering that whenever two classes or notions are said to be roughly distinct, that means that they are partly distinct and partly not so. The ends of the scale are plainly separate, like the snow and the water, but a middle portion is inde- terminate — cannot be said to belong exclusively to either end of the scale. In this connection we may notice in passing that the epithet ' so-called ' can often be turned to useful account in justifying an apparent self-con- tradiction. When anyone objects that so and so (A) cannot possibly become something else (non- A) by any process of gradual change — for example, that ' man ' cannot ever have been less than human, or that unconscious experience cannot ' become ' conscious ^ — the reflection that the actual things or cases in question are only so-called A and non-A, that these names are perhaps inexact, may often be of service. We are all too ready to see in words a mysterious datum behind which it is impossible to go. It takes a long apprenticeship to realities before we begin to get free from this ' Cf. Green's Prolegomena to Ethics, pp. 18-20. See also p. 215 of this book. THE NATURE OF ROUGH DISTINCTION 21 illusion. But there is small excuse for anyone nowa- days remaining ignorant of the fact that Nature is full of examples of a development which appears in our clumsy and rigid language as self-contradiction. Every child that outgrows childhood, every seed or germ that becomes other than seed or germ, every fact that changes its character in the least degree, proves to us daily that the ' Laws of Thought,' ' those pillars of elementary logic, are too ideal and abstract to be interpreted as referring to the actual things or particular cases that names are supposed to denote. § 6. Some other ways of characterising rough distinctions. — We need not insist on using the epithet ' rough ' or ' artificially sharpened ' to the exclusion of all other possible ways of describing the fault to which distinctions are liable. Indeed, recognition of the fault is so common, and so plainly important, that everyday language has thrown up many other forms of expressing the same meaning. We often speak of ' broad ' or ' loose ' distinctions, or ' fluid ' or ' slippery,' and philosophers sometimes call them 'abstract,' or object that common-sense ' takes too abstract a view of the facts ' ; sometimes we say that a given distinction 'must not be taken quite strictly,' or ' See p. 71, note. DISTINCTION ' will not bear pressure,' or ' is clear on paper but not in real life ' ; or we recognise cases where ' the facts are more complicated than the names.' And besides these that come first to hand, there are others in common use, but we may be content to make their acquaintance gradually. § 7. Unreal distinctness. — Another convenient phrase for roughness of distinction is ' unreal dis- tinctness.' As this name is meant to suggest, it is by contrast with (what are taken for) realities ' that words are seen to imply unduly abrupt dis- tinctions ; the discontinuity of language is noticed only so far as we think we see the real continuity of Nature. And this phrase ' unreal distinctness ' will often be used in the following pages for the fault to which names and distinctions are so liable. Some further explanation of it is to be found in the chapter on Unreality. § 8. The applicability of distinctions. — But a still more useful phrase is the 'applicability' of names and distinctions ; a name being vague, or a distinction rough, so far as it is inapplicable in concrete cases. That, in fact, is the reason why unreal distinctness matters. Ideally, there is never any difficulty about a distinction ; whatever diffi- culty there is attaches only to its application. Take, ' See Chap. v. pp. 58-70. THE- NATURE OF ROUGH DISTINCTION 23 for instance, the notion of truth, the distinction between truth and falsity. Everyone knows, of course, what he means by calling a statement true, but the difficulty is to sum up the distinguishing characteristics of truth, as we might sum up those of the elephant or the gorilla, so as to enable actual specimens to be identified by means of the description. The question ' What is truth ? ' can be answered easily if, like Mr. Chadband's audience, we are content with a merely ideal answer.' In idea, truth and falsity are distinguished by a line of absolute sharpness ; where either begins the other ends, without an intermediate region. It is when we come to apply the distinction to actual ' To express the background of a name by tacking to that name •a negative prefix (e.g., organic, ZKi7?-^ia;»zV)ismerely a way of shirking the question how the line is to be actually drawn. It leaves the distinction ' ideal,' without attempting even to suggest the actual particulars of the contrast. Often when the sense in which a word is used lacks positiveness— that is to say, when the actual particulars intended to be referred to are left doubtful — help can be given by contrasting the word with some other positive notion which happens to be a little less indistinctly understood. Difficult notions, such as liberty, or justice, or discretion, can occasionally have some false application disclaimed by contrasting them with one opposite rather than another — say with slavery, partiality, rashness, instead of with restraint, or mercy, or valour ; and in plenty of pairs, like life and death, love and hate, health and disease, pain and pleasure, each of the opposite notions helps to interpret the other, by supplying our thoughts vdth more or less of remembered actual detail. But this aid to interpretation is lost so far as one of the pair of names is merely the ' formal negative ' of the other. DISTINCTION cases that all the uncertainty is discovered. On which side are we to class a truth that has some error mingled with it, or an error that contains a certain amount of truth ? And when does the claim that a given truth or a given error is pure and simple cease to be doubtful ? Truth, we are solemnly told, is ' conformity of knowledge with its object.' No doubt it is ; but when does knowledge so conform } Thus the enquiry after a definite meaning is always the enquiry how some general name shall be applied to particular cases ; whether such and such a case would or would not be properly classed as ' A.' Whatever special form the enquiry may take, and however little express reference may be made to particular cases, such is always the central purpose of the request for a definition. The truth of this statement might be tested by examining an infinite number of instances in detail ; but it may also be verified more shortly by noticing what happens when, in answer to the demand, an un- satisfactory definition is given. From a philoso- phical, as distinct from a merely literary, point of view, there is only one kind of unsatisfactoriness that definitions ever suffer from; — failure to tell us, after all, exactly what things or cases are meant by the name defined. The process of defining, as THE NATURE OF ROUGH DISTINCTION 25 everyone knows, is the substitution of one word (or several) for another ; and, except for the sake of following custom, we need only take care that .^he substituted word is less hard to interpret than the word for which it was substituted. The enquiry after a definition is an enquiry after meaning ; and meaning is, in the end, always interpretation into fact. If a word, B, ever serves to explain the meaning of another word, A, that is only because B is more easily interpreted into facts. For words are not valuable in themselves, but only as refer- ring to something beyond them. The demand for an applicable definition, the attempt to make an assertor translate out of the abstract or ideal into the concrete or actual, to force him to compare the solid facts of the case with the airy notions which are supposed to refer to them, is a plan of controversial attack that is probably as old as controversy itself. In the hands of Socrates it attained to the dignity of a ' method,' and, though less entirely trusted now, and made somewhat less rigid in application, it remains in effective use at the present day. Every reader will remember plenty of cases where the matter-of- fact enquiry about some difficult notion— justice, humour, luxury, culture- has been put forward with telling effect, justly or unjustly. ' Give us a 26 DISTINCTION definition that we can really apply to the facts.' It is a demand identical in purpose with the request for ' a single definite instance ' in support of a sweeping generalisation ; and, as said in the pre- ceding chapter, one of the problems we have to discuss in the present book is, how to hold the balance fairly between this very plausible demand for ' definite ' language, and the best excuses that can be made for vagueness. § 9. General and particular grounds of application. It is hardly necessary perhaps to mention, except for the sake of completeness, that a name or dis- tinction may lack applicability in two different ways : through our ignorance of the general grounds for its application, or through our ignorance of the facts of the concrete case to which we propose to apply it. However clearly, for instance, we know what a hero means, we may be in doubt whether Parnell was a hero ; and however well we know Stanley and his actions, any flaw in our general notion of a hero may render it doubtful whether the name should be applied to him. To apply correctly a name like hero (or a distinction like that between hero and charlatan) demands an intimate knowledge of facts that are often misleading when superficially seen. Boulanger's valet may know more of such facts, or even less of them, than the editor of an THE NATURE OF ROUGH DISTINCTION 27 English newspaper. Always our power of apply- ing a distinction to actual cases is limited by our knowledge of the real nature of those cases, and as our knowledge grows our applications will alter. To apply always with perfect correctness the dis- tinction between truth and falsity would need an omniscient mind. § 10. The ideal and the actual. — But the most useful of all the notions which help us to under- stand the subject generally is one for which I can find no better name than the distinction between the ideal and the actual. As already said, there is never any difficulty about a distinction except the difficulty of applying it. The abstract form, or idea, of a distinction never alters ; the only change that comes, with our changing moods or our grow- ing knowledge, is in the details of application, the way we interpret the notions contrasted, the kind of things that we call A and non-A respectively. Life and death, good and evil, truth and falsity — the application of such names may change, but never the fact of contrast, never the need for find- ing some distinct and opposed meaning for them. If we ever say, for instance, ' Evil, be thou my good,' it is not the abstract distinction that we blur, but the concrete application ; if evil is to be our good> something else must be our evil. 28 DISTINCTION As an easy example of the uses to which this notion and phrase may be put, take the case of a distinction like that between animal and vegetable. Every naturalist allows that certain organisms may,, for all he can say, be classed just as correctly on one side of the line as on the other. Do these organisms, then, belong in fact to an intermediate class, or must they be in fact either animal or vege- table, one or the other, although in our ignorance of the facts we cannot class them definitely ? Such a question will not appear a serious puzzle to any- one who keeps clearly separate the ideal and. the actual aspects of a distinction. The distinction between animal and vegetable may or may not be ideally sharp ; that depends on whether we choose to make it so. If we decide to recognise an inter- mediate class, the distinction ceases to be ideally sharp ; and only if it is ideally sharp can we have any right to say that a doubtful organism must actually belong to one or the other of the two opposite classes. The ' must,' in short, depends upon a purely arbitrary arrangement of words and meanings. In the case of distinctions like hot and cold, or old and young, we do not commonly make the line ideally sharp ; in the case of distinctions like straight and crooked, or true and false, we do make it so ; and in the case of distinctions like animal THE NATURE OF ROUGH DISTINCTION 29 and vegetable, or man and beast, custom appears to be at present a little uncertain. The question whether ' animal ' and ' vegetable ' are distinct or overlapping classes is answerable precisely in the same way as the question whether fox-terriers are a ■distinct breed or a mixture of two or more. The answer depends on our own choice of a definition. What makes this double aspect of distinctions important is the fact that all the more lively and lasting disputes in the world are due to the conflict of rival ideals whose actual interpretation is diffi- cult. And by shortly reviewing' the nature of these more permanent sources of controversy we may now begin to raise the question as to the extent of rough distinction that exists, and as to its influence on thought and opinion generally. But first there is one example of rough distinction that specially demands some notice — the distinc- tion between philosophy and common-sense. ' Page 41. CHAPTER III PHILOSOPHY AND COMMON-SENSE' Since the difficulty of ' drawing the line ' is felt in common life as well as in the most abstruse philo- sophy, and since the attempt to solve that diffi- culty is in its very nature more or less philosophi- cal, it would be useful in any case, to explain the way in which the relation between philosophy and common-sense is here throughout conceived ; and there is all the more advantage in doing so since the contrast in question forms a typical instance of unreal distinctness. For if we are candid we must confess that we do not very clearly know what, after all, philosophy is ; we cannot, at any rate, give an ' applicable ' definition of it. In a vague way we all know, of course, what is meant by the term. Our views are commonly said to be philosophical in so far as they are comprehensive, many-sided, remote from ' The philosophical reader is warned not to expect in this chapter any reference to Reid and his followers. PHILOSOPHY AND COMMON-SENSE 31 what is sordid, grovelling, or ' practical ' in the lowest sense. The questions that are specially- called philosophical give themselves out as being the largest and loftiest questions that human curiosity can raise — questions like that of the existence of God, the immortality of the soul, free will, the relation of man to the universe generally, and similar serious topics. Or, again, there are certain illustrious names — Plato and Aristotle, for example — names of men whom all agree to call the great philosophers ; and we may study their works until we are able to say a great deal about the earlier history of most of our modern problems. Or, again, we establish professorial chairs of philo- sophy, and appoint to them men who will undertake to lecture in various named departments of study, such as logic, psychology, and ethics. Yet we do not arrive at an applicable definition. The question remains unanswered, what it is that really constitutes this subject and method as dis- tinguished, say, from science or from common- sense. If the name ' philosophy ' is to be given to all the most rational .attempts to answer certain questions that we agree to call philosophical, still we are left asking why, then, these questions specially deserve the name ; and the same holds of the plan of naming the subject from the writers or 32 DISTINCTION from the special fields of study : we then have to enquire why only certain men deserve to be called the philosophers, or what makes logic, psychology, ethics, &c., departments of philosophy. Nor is the decision by means of the comprehensiveness or the loftiness of the views much more satisfactory. At least, if this be made the decisive fact, it is hard to see why, in modern times, we have left off calling physics, or even astronomy, philosophical subjects, and have shown some tendency to include political economy in our philosophical courses. And clearly it is not easy to say at what point views begin to be comprehensive or lofty. When the term ' philosophy ' is understood in the light of its everyday applications — that is to say, when our notion of its meaning is derived merely from our knowledge of the questions or the people or the books, and so on, that are commonly called philosophical — it is hardly surprising that some dis- inclination should be felt for so apparently hopeless a pursuit. As everybody can see for himself at a glance, authority in philosophy there is none, or none that is not authoritatively disputed. If you study any branch of science you find an increasing mass of knowledge conquered, a fairly constant progress both for the scientific and for the unscien- tific ; there authority has a meaning. But if you PHILOSOPHY AND COMMON-SENSE 33 study philosophy, it seems that you either follow some teacher and belong to a ' school,' or at least are almost sure to be suspected of that narrowness. In the most favourable case you can only create a limited following of your own. For a time, perhaps, you and your disciples will occupy a half-way position between two schools, distrusted by both. And when these disappear, some other school will have established itself in direct opposition to yours, and will so continue until a new temporary reconciliation is effected and the old round begins again. The history of philosophy can indeed be taught, and can best be learnt, by those who are fonder of antiquarian studies than of direct attempts to answer the questions raised. So that, in the end, to be a philosopher is either to fancy your answers more final than they really are, or else to know what other philosophers have wrongly said, and yet not to be able yourself to say it rightly. So unfriendly a statement of the case, however, is of service rather in making excuses for declining to take any trouble with the subject than in dis- turbing those who have already begun to take some trouble. These latter would call it a carica- ture. They would rather say that philosophy is an ideal that may be approached, although perhaps D 34 DISTINCTION no actual person or doctrine exactly reaches it ; which explanation implies that no such thing as philosophy exists in the world, as distinct from science or even from common-sense, but that all our knowledge is more or less philosophical, and therefore also more or less unphilosophical. And the deplorable want of finality, they would say, in the actual doctrines of ' philosophy ' is a result of their partial failure, not of their partial success, in reaching the ideal. It is not because of philosophy, but in spite of it, that we are still to some extent puzzling over the oldest problems. We all agree that there exists some difference, some reality of contrast, between philosophy and common-sense. But the sharpness of the contrast is, doubtless, very variously conceived. There are people who, in the name of philosophy, appear to aim at discarding common-sense altogether ; others, in the name of common-sense, succeed to a great extent in avoiding all views that are noticeably refined or difficult ; and others, again, occupy a position somewhere between the two extremes, attracted now by the delights of visionary insight, now by the notion of sanity, repelled now by the fear of getting lost in cloudland, now by dislike of the clumsy roughness of ordinary ways of thought. And it is through sympathy with these latter — PHILOSOPHY AND COMMON-SENSE 35 perhaps the most numerous class among those who have time to think at all — that we shall best learn how artificial the contrast is. Our own experience of changing moods in regard to refinement of insight will teach us more forcibly than anything else can that philosophy and common-sense cannot in real life be entirely severed, that the sublime and the ridiculous, the sane and the stupid, lie near together ; that the two opposed ideals, like all other opposed ideals, have each their practical limitations. The view here taken, then, is that philosophy is only common-sense with leisure to push enquiry further than usual, while common-sense is only philosophy somewhat hurried and hardened by practical needs. As an ideal, philosophy is cha- racterised by freedom from assumptions. The attempt to criticise assumptions to the utmost is the distinguishing feature of what has always been recognised as the philosophical spirit. Just where science leaves off questioning, philosophy (as dis- tinct from science) begins. But the philosophical spirit, or impulse, is only one among others that drive us on, and the opposite impulse of common- sense inclines us to dread wasting time in doubts and refinements that are useful only on rare occa- sions. Still more than science, philosophy feels an D 2 36 DISTINCTION interest in exceptions as such ; while common- sense, like its feebler cousin, ' worldly wisdom,' shuns the exception and loves the general rule. Philosophy, so far as it is at war with common- sense, seeks to amend faulty rules by dwelling on their exceptions ; its business is to justify truths that to common-sense are foolish or dangerous paradoxes. The business of common-sense is to justify the broader aspects of truth, to be ready though rough, to be right in most cases at the cost of being wrong in a few. And, further, we may say, perhaps, that the difference between a more and a less philosophical view of things consists in greater or less acquaintance with, and readiness to admit intelligently, certain notions which are paradoxes to the commoner kinds (at least) of common-sense. Take, for instance, the assumption so widely made that a cause, as such, is prior to its own effect ; the denial of this is a paradox of the kind referred to. Here and there, no doubt, the admission is made by common-sense, in a rather half-hearted way, that there are some cases where ' it seems ' that cause and effect are simultaneous, or cases where a relation of mutual efficacy ' seems ' to exist between two phenomena— say panic and danger — or where something seems, if not exactly to cause itself, at any rate to hasten its own growth or advancement, PHILOSOPHY AND COMMON-SENSE yj so that, for instance, ' nothing succeeds like suc- cess.' But these cases are not, as a rule, taken quite seriously ; they are regarded either as mere illusions, or as verbal quibbles or epigrams. Common- sense feels sure that the facts are not exactly as the words would make them appear ; that if there is any difficulty in applying our ordinary notion of cause and effect to them, then some way out of that difficulty could surely be found if the enquiry were really worth the trouble ; and meanwhile we treat the phrases lightly, and remain secure of the axiom that a cause, as such, must come before its effect. And so, if the doubt be raised whether philo- sophy can tell common-sense anything important about rough distinction, we may answer that in the first place we are not compelled to admit any such thoroughgoing contrast as the question assumes to exist between them. The interesting question is, whether or no our ordinary tact in the use of distinctions can be improved at all by generalising and arranging our knowledge of the subject. The existence of this ordinary tact need not be denied ; all that need be denied is the pretension that the light of Nature suffices for all of us on all occasions where words and distinctions are used — that in the use of language we are all infallible. Genuine 38 DISTINCTION common-sense has a stronger case than this to rely on, namely the practical need for acting sometimes in a hurry. And in the same way philosophy has a stronger case when it ceases to set itself up as distinct from common-sense and above it. Any philosophy which is not bent on being merely destructive, or which has even distantly recognised the aims and the power of the historical method, knows very well that common-sense will never show itself to have been wholly irrational. What common-sense has seen and believed, there have been some practical reasons for seeing and believing ; some practical purpose, however limited, has been effected by means of even its most illusory notions. The faultiness of its method is only one side or aspect of a process which has had at least twO' sides, two aspects, each worthy to be considered ; a process in which two conflicting aims, two sets of data, are to be taken into account. And the business of philosophy is not to content itself with railing at the unphilosophical, but rather to learn from common-sense how to correct the hard and narrow judgments which at first arise out of the abstract method which a philosophical enquiry involves. Its best work will be done as common- sense itself attempting to guard, by the method of PHILOSOPHY AND COMMON-SENSE 39 reflection, against the errors that belong to the readiness and rapidity of its ordinary judgments. Some of the short cuts of common-sense are useful in a large majority of cases and misleading in a small minority. The interesting problem is how to discriminate between the cases that belong to the one and to the other of these classes. There is less harm in conceiving of common- sense as existing in various grades of approach to perfection of wisdom, and of the higher grades as being able to help the lower to generalise the sub- ject of distinction. This conception does not, at any rate, compel us to class ourselves as philo- sophers on the one hand, and common-sense people on the other ; it does not even compel us to separate people as possessing higher common- sense and lower. Within each of us there is a higher and a lower common-sense, a more and a less hurried method of forming judgments, and whatever we do in the way of improving our shallowest views upon any subject only results in winning new ground for common-sense to occupy, not in achieving a victory over common-sense itself. And in spite of theunsatisfactoriness of the contrast between what are only higher and lower grades of the same quality, one thing at least is clear — that, whether in the name of philosophy or 40 DISTINCTION not, we are here concerned with the war between two methods sharply opposed in idea though not in their actual manifestations : the method which loves a short cut, and the method which aims at taking the utmost care. When this fact is clearly seen we may speak of ' common-sense ' and ' philo- sophy ' without much fear of being misled by the unreality of the contrast. We shall find some convenience throughout this book in speaking as if the contrast were really sharp, but the reader is hereby notified that any such manner of speaking is merely a harmless device of expression, adopted for the sake of avoiding heavier phrases. CHAPTER IV RIVAL IDEALS Anyone who remembers what is the kind of dis- putation he has most frequently heard, or shared in, will understand what is meant by our saying that questions of fact have less vitality, less con- troversial aptitude, than any other kind of question. Of course we need not in saying this assume that the distinction between matters of fact and matters of theory or opinion can be drawn with exactness. That is another example of ' unreal ' (or ' ideal ') opposition. Like plenty of other distinctions in daily use, it is serviceable only so long as we do not take it very strictly. But at any rate the disputes which are least easily settled lie at the opposite end of the scale from ' questions of fact.' For we commonly mean by a question of fact a question which is open to settlement by methods of verifi- cation which no one but a desperate sceptic can refuse to accept as final. And the desperate 42 DISTINCTION sceptic is hardly to be taken into account when we are trying to put a workable meaning into the phrases that everyone has to use. ' About matters of taste there is no disputing/ Rather, disputes on all other subjects are mild and transient by comparison. But the proverb never- theless expresses a truth — namely, that where a dispute can be easily seen to turn upon some un- avoidable personal difference of inclination in the disputants, they can reach an agreement to differ ;, that is to say, the argument may then be brought at once to its only possible end. So far as the proverb is false, its falsity is chiefly due to the fact that differences of taste so often disguise their real nature ; questions of taste in food we are accus- tomed not to argue ; questions of taste in dress, or in pictures, or in people, provide rather more de- batable matter ; and our widely different views in morals, religion, philosophy, are seldom described as matters of taste at all. Bad taste in matters where reason can be appealed to is supposed to be open to some correction by the methods of reason. But we certainly need not make any point of using, for the more serious kinds of dispute, so undignified a name as questions of taste. It will serve all purposes equally well to describe them as H/VAL WEALS 43 cases of the conflict of ideals ; and indeed the latter name is really more wide-reaching. The notion of an ' ideal ' may be fairly used to include all the less grossly material inclinations, from fancy in purely artistic matters up to choice in the most delicate shades of thought and conduct. Will anyone maintain that in art-criticism, taste is wholly irrational ? At any rate we have each our own opinion as to the merits of certain novels, or plays, or pictures, and where the opinions conflict general grounds are often appealed to for justifica- tion of the views we have taken. We are not always content, that is, to regard our artistic likings and aversions as mere ' matters of taste,' wayward and accidental, but are often tempted to see that other people's tastes, when different from our own, are spoilt by some defect in their educa- tion or powers ; which defect we are more or less ready to characterise and explain. Such and such a work of art appeals, we say, to the gallery, or to the savage within us, or to the spirit of sentiment- ality, or .whatever the damning phrase may be; and, on the other side, he who cannot perceive its excellence is said to be deficient in sympathy, or in sense of humour, or in some other quality that everyone claims to possess. Indeed, it is easy to generalise in this matter so as to convince ourselves 44 DISTINCTION and irritate other people — a state of things out of which a discussion will often arise. Throughout the class of discussions where a conflict of ideals lies at the root of the difference of opinion, there is always ' much to be said on both sides ' ; which is doubtless precisely the reason why this kind of dispute is the longest lived and the most unappeasable. These are the questions that arise afresh in every generation, the questions whose final settlement seems for ever just to elude our grasp. And even the most peaceable life is full of such conflicts. Shall we put our trust in Conservatism or Reform ? Should we prefer old books or new ? Ought fiction to edify or merely to please? Are we to condemn ' theoretical ' people as inexperienced and unpractical, or to condemn the ' practical' people as unenlightened and narrow ? Shall we philosophise, or shall we be content with common-sense .' Perhaps these few examples will sufficiently bring before us the kind of conflicts ot opinion that occur most often, that last the longest, and that stir up the angriest feeling, and drive the deepest wedge between the opposite parties. But now as to the general characteristics of these ' ideal ' disputes. In the first place it is naturally the more deniable truths that we com- monly find as a matter of fact denied. The RIVAL IDEALS 45 extreme cases of certainty are not, as a rule, those which lead us to separate into hostile camps, but rather the cases where some vague faith of our own is attacked effectively ; it is, therefore, rather our shaky beliefs than our perfectly firm beliefs which provoke us to enter upon their defence. And, further, it is very instructive to notice the part which controversy plays in hardening these semi- beliefs into professed certainties, and the way in which, in the heat of discussion, we fortify our own views by making the most of their difference from those of the opposite party. All controversy, as such, implies a sharp division between the opposite sides — the ' ayes ' and the ' noes.' And precisely those controversies which have most vitality arise where (since there really is much to be said on both sides) the spirit of partisanship stands most in need of other support than can be given by pure reason or regard for truth. For we often cannot properly despise our opponent's opinions until we have first caricatured them slightly, and our own will never be the worse for being slightly idealised.' We are Conservatives, for instance, not because we are stupid or timid, but because we are wise enough to ' In a chapter on ' Rival Ideals ' I must apologise for using the word idealise in this more popular sense. But there really seems to be no other word to use — unless glorify commends itself to the reader. 46 DISTINCTION distrust all abstract system-making in politics, or the fanciful inexperience which imagines that the millennium can be suddenly brought about by revolution or by Acts of Parliament. We are Radicals, not because we are rash and ignorant, or devoured by envy, but because the Conservative is essentially a slug, and because the course of history shows that the old order is always changing, and that all progress involves a certain amount of rough experimentation and the running of certain risks. Our own Conservatism, or our own Radicalism, is, we assume, faultless, or is free, at least, from the excesses which our opponents wrongly suppose to be essential to it. And if we look at any of the great battle-grounds of opinion we find the same tendency to exaggeration of the contrast, the same practice of idealisation and caricature. If the instances already suggested fail by chance to remind us of actual controversies, we may look for ex- ample at the dispute that raged some years ago between Mr. Harrison and Mr. Matthew Arnold about 'culture,' Mr. Harrison complaining in effect that Mr. Arnold had idealised the notion out of all resemblance to the actual facts of the world, and Mr. Arnold complaining that Mr. Harri- son's notion of it was little more than an ignorant caricature. RIVAL IDEALS 47 In all those cases where ideals are brought into conflict it is the old rivalry between the abstract and the concrete meaning, the ideal and the actual, that chiefly keeps the dispute alive. The prophet or preacher of some ideal — culture, for instance — proceeds smoothly enough so long as he preserves the oracular tone which deals in safe generalities. The praise of culture, like the praise of goodness, may be sung in a vague and soothing manner -which, just because it recommends no definite labour and holds up to scorn none of our obvious actual habits, leaves us serenely convinced of edifi- cation. The oracular form enables us to interpret the doctrine exactly as we please ; the preacher seems to be talking to friends in a friendly way and using his words in the ' fluid and passing ' ' manner that always renders intercourse easy. It is delight- ful to be thus excused from the trouble of defining exactly what we mean. But no prophet or preacher can long remain quite vague and negative, or complaisant to the whole of the world around him. Sooner or later he must run counter to prejudice ; being human, sooner ' See Literature and Dogma, chap. i. Mr. Arnold knows very well how to remind us that a given word is ' by no means a term of science or exact knowledge, but a term of eloquence, a term thrown eut, so to speak, at a not fully grasped object of the speaker's con- sciousness — a literary terra, in short.' 48 DISTINCTION or later he will show his enmity to something defi- nite and actual. And then it is that doubts arise as to the exact interpretation of his ideal. Let him assert, for instance, that the pursuit of culture is destined to ' transform and govern ' ' religion, and he will find his hearers divided into those who do and those who do not proceed to enquire more critically what, after all, is meant by culture- Religious people are in arms at once ; they begin to see some difference between ideal and actual ' culture.' If the prophet tells them that culture is ' a study of perfection,' ^ and that ' it moves by the force not merely or primarily of the scientific passion for pure knowledge, but of the moral and social passion for doing good,' they are apt to reply that this would surely be a better description of reli- gion itself than of culture — a name which suggests something much less active and much less wide in its sympathies ; of culture as it actually exists they would say that its moral and philanthropic passion is rather languid, and that its attitude towards the many is decidedly pharisaical. If ' culture ' means anything definite at all, they begin to ask, who are the actual people who may be said to possess it .■• Are they or are they not members of the small and sickly cliques who pride themselves upon their ex- ' Culture and Anarchy, chap. i. " Ibid. RIVAL IDEALS 49 cellent taste in books or china— the men whom Mr. Harrison calls, in regard to questions of manly interest, ' the poorest mortals alive ' ? Is it, for instance, some armchair dilettante who proposes to ' transform and govern ' the spirit that ruled the life of Gordon, or Father Damien ? The distinction between culture and its opposite is fairly typical of the difficulty so often raised by the sharp opposition of ideals. The formation of these abstractions, the analysis of concrete fact into separate elements which are never found quite de- tached and pure, is at once the foundation of the general method of science and one of the most dangerous of all the sources of illusion. When we preach an ideal we are apt to interpret our abstract notion in some obvious concrete sense until the statements we have made are found to be untenable, and then to retire into the mists of abstract meaning until the attack blows over. This may , be illustrated in the case we are here considering. In spite of the claim that the aim at culture is simply the general aim at perfection, there are certainly passages in Mr. Arnold's account of what culture is which go far to justify his opponents' idea of its actual meaning. For instance, on one occasion ' he finds that an American writer, fired by the ' Ibid., chap. ii. E so DISTINCTION belief that industrialism, or material progress, is the chief means of ' making reason and the will of God prevail,' proposes to set up the class of indus- trialists as typical men of culture. And this, says Mr. Arnold, 'is undoubtedly specious ; but I must remark that the culture of which I talked was an endeavour to come at reason and the will of God by means of reading, observing, and thinking ; and that whoever calls anything else culture may indeed call it so if he likes, but then he talks of something quite different from what I talked of.' Reading, observing, thinking ! That begins to be a more definite account of the actual method of culture ; but why, then, should Mr. Harrison be blamed for supposing 'culture' rather a narrow conception of the aim at a perfect life? Is the life of a reader or thinker, in truth, ' the ideal of a human perfection complete on all sides ' ? Are not books, too, a form of ' machinery ' } Or, as a matter of fact, are students and literary people specially the class that possess ' a finely-tempered nature' .'' The more we consider the actual topics of dis- cussion, the more we shall find that, while the conflict of ideals is what gives to such topics their chief vitality, the doubt how to apply abstract notions in concrete cases is what chiefly keeps the ideals in RIVAL IDEALS ji conflict. The life of controversy depends partly on the difficulty of verification, and in these vague and wordy discussions the difficulty of verification is at its highest. It depends also partly on the extent to which matters of taste can disguise them- selves as matters of more general import, and the vaguer the discussion, the easier is this disguise. But it depends most of all upon the power which vagueness of language gives us of putting a better or worse construction on what ourselves and our opponents really mean. The special form the discussion takes is not, indeed, always so simple and clear as in the example we have just been noticing. The point at. issue is not always on the surface, the merits and demerits of this or that quality or line of conduct. Quite as often the dispute pretends to be aimed throughout merely at the meaning of a word or at the drawing of some fine distinction : what is genius} for example ; or how to distinguish correctly between fancy and imagination, or between legitimate speculation and gambling ; whether right is only might in the end, and similar forms of apparently verbal problems. Yet the real nature of such discussions may easily be seen to be ethical or ideal ; what gives them their chief, importance and interest is the relative attractiveness or repulsiveness of the notions, the S2 DISTINCTION implication of comparative praise or contempt that underlies the names. Genius, whatever it may be, is something we revere wherever we think we find it ; the distinction between fancy and imagination is motived by the wish to praise the one at the expense of the other ; ' while in the case of speculation and gambling, or right and might, the ethical purpose of the contrast is still more plain to see. Such distinctions are thinly disguised forms of the conflict between opposed ideals. It is sometimes hinted by practical men that all philosophy is mere word-spinning, mere dis- puting about the meaning of words. And some philosophers^ themselves would perhaps have admitted the charge, in a sense of their own. Like most other philosophical opinions, it may correspond to much or little insight ; at any rate, it need not mean simply what the unphilosophical intend by it, that philosophy is altogether trivial. In the light of the common examples just brought forward it is plain that disputes about words may have a practical justification as referring, though ' ' Such distinctions,' says Mr. Ruskin, in disparaging the im- portance of this particular discussion, ' are scarcely more than varieties of courtesy or dignity in the use of words. ' But if they beeven so much as this it is enough for our purpose. ' E.g. Condillac, who said that une science bien trait ee n'est quUine langue bienfaite. RIVAL IDEALS 53 obscurely, to ideals of conduct or general choice. And not only is it impossible to draw a sharp line between the class of controversies we have been discussing and those which may be called in the technical sense philosophical, but the latter consti- tute rather the extreme case of what is essential to the former. ' Philosophical ' discussion is less matter-of-fact than any other, and is less easily brought to an end ; it is most of all affected by the vacillation between ideal and actual meanings ; and, equally with the great politicians and divines, the great philosophers are apt to exaggerate the difference between rival systems by means of idealisation and caricature. It is so easy to put either a good or a bad construction upon any formal doctrines; so difficult to avoid this sort of un- fairness. And further, the conflict of ideals is plainly ex- emplified in all the chief philosophical debating- grounds, however technical the language in which the discussions are carried on. To a great extent philosophy has always been concerned with ethics ; that is to say, with the attempt to harmonise con- flicting ideals of conduct : and even in its least directly practical questions, such as the ever- recurring puzzles about the nature of truth or certainty, or about the ultimate foundation of 54 DISTINCTION things in general, the problem still is to harmonise opposed ideals, though here they are rather ideals of intellect than of character. For instance, one of the great dividing forces in philosophy at all times has been the rivalry between two opposite methods of general explanation — that which explains small things by great ones, the part by the whole, the; many by the one {e.g. all earthly facts as related to their one cause and substance) ; and that which explains great things by small ones, the the whole by its parts, the one by the many {e.g. the system of Nature as a ' concourse of atoms '). And behind all lesser intellectual differences lies that between the temper which easily believes in order to, understand, and the temper for which belief without understanding is hardly possible. We must look more closely at one especially of these deeper, sources of endless conflict in a later chapter.' At present enough has been said, I hope, to help us further in seeing why we should raise our general questions as to distinction and definition. Keeping in mind the few examples already given, we can now begin to discuss these theoretical questions with some insight into the reasons why the enquiry is thought to be worth while. ' Page 202. RIVAL IDEALS 55 It appears, then, that in all the more lively and lasting disputes some ideal, some ' notion,' is con- trasted either with one, or with several, others. To a great extent the vitality of discussion in general depends upon the absence of sharp definition, sharp distinction ; or, rather, upon the absence of that kind of sharp distinction which is applicable, not only to the notions themselves, but to the actual facts to which they pretend to refer. The less definite the words which we are forced to use in describing our aims or beliefs, the more the actual difference between the opposed aims or beliefs lies open to exaggeration by means of idealisation and caricature. For instance, were the line between ' culture ' and its opposites firmly drawn, were the notion of culture marked out by a definition un- mistakably applicable in every actual case, the fire of the dispute would be deprived of its fuel, since then we should no longer be able to connect the name 'cultured' only with those we admire or despise ; the cultured would then take their place as a mixed company, like peers, or women, or Frenchmen, or any other definite class. I do not, of course, deny that rash general assertions are sometimes made about peers, or women, or Frenchmen, but I hope the reader will agree that such assertions are S6 DISTINCTION mostly to be explained as carelessly expressed, their real reference being only indirectly to the individuals, and directly to vague ideals which are very obscurely suggested — for example, it is ' Gallic lightness ' that we admire or detest, rather than every actual Frenchman ; and that where no such explanation can be found, all sensible people are agreed that the assertion is foolish. Surely no definite class of people resemble each other so closely that all the members exemplify exactly any one ideal ? Surely not all actual peers, or women, or Frenchmen, are equally attractive or repulsive ? At least, if anyone really finds them so, we must leave him out of account. The question is not an easy one. How far it is necessary to apply abstract notions in actual cases in order to give them a definite meaning. By the light of Nature we are very likely to misunder- stand it, and if we rely on elementary logic we are perfectly certain to do so. For the 'laws of thought,' though ideally true, are false in every case as applied to actual things.' Yet excessive belief in the 'laws of thought' is not wholly con- fined to beginners in logic, but all higher logic, whether intuitive or elaborate, demands a constant ' See p. 71. Jl/VAL /DEALS 57 watchfulness against the early superstition. This matter will, therefore, require a careful discussion, the outcome of which, I hope, will be a fuller knowledge of the rights of the demand for an applicable definition. CHAPTER V UNREALITY Our frequent use of the word ' unreal ' will require, sooner or later, some explanation and apology, and the point now reached appears to be a good one for the purpose. The reader can hardly fail to see by this time how much is made to turn upon the distinction between real and unreal in our whole view of the subject ; and since this distinction has always been one of the quicksands of philosophy, some care will have to be exercised in picking our way across it. The best ' short account of the manner in which the word ' unreal ' is here throughout employed, ig that it is expressly intended not to imply an answer to the metaphysical question. The metaphysical question about reality is how to distinguish real (or absolute) reality^ from the reality which is only ' Or best available at the present stage of our discussion. After p. 1 80 we can call it simply a Reference-name. ' The phrase ' real reality' is of course open to verbal criticism ; UNREALITY 59 apparent (or relative) ; and whatever views I may hold on this extremely difficult subject, either knowingly or otherwise, I do not wish to inflict them on the reader. Everyone is driven, at times, to question the reality of things that seem most real to someone else. Such questions are more €asy to raise than to answer finally. No doubt the reader has his own opinions as to which of our so-called realities most truly deserve the name ; and if his view should differ from mine, who is to judge between us, in the end .' Not I, at any rate ; nor do I see how I could accept his view, in the very act of disputing it. With the best inten- tions, one cannot be both judge and party in a suit. And so it is only admitted reality that we are here to take into account ; we shall use the terms real and unreal only so far as the reader and I can agree on their application. In practical life we are all in the habit of. admitting, doubting, dis- puting, denying, the ' reality ' of the things that we and our neighbours talk about ; and though we do not all agree to call the same things real^for tut it need neither mislead nor oflfend anyone. It is here used only to express the fact that the metaphysician, as such, is inclined to ijuestion the reality of ' things ' which his less metaphysical neigh- bours accept as real. His function is to distinguish (so-called) realities as real and unreal. I do not see any better short way of expressing the same meaning. 6o DISTINCTION instance, grown-up people think the troubles of childhood often only fancied — yet a certain amount of agreement is reached as a matter of fact. Those engaged in physical science agree on the whole as to what are realities of observation, and unscientific opinion is much more easily satisfied. There are people who agree with each other that ghosts are real. For our purposes, the most interesting fact about such agreement in general is that it often makes no pretence of being more than provisional. Just as we may admit the truth of some doubtful assertion ' for the sake of argument,' so we often admit, for a passing purpose, the reality of some- thing which on other occasions we should be dis- posed to call unreal. For instance, future and problematical occurrences are often named as sub-' stantives, in order that we may speak of them as if they were present or actual. The symboJs used in mathematics afford perhaps as simple an instance of this as can be found. When we say that 2 + 2 = 4, we are only expressing concisely the fact, that «y we take two units of anything, and add two more units to them, we shall have the same amount as if we had taken four units at once. Similarly, a fraction, or a set of figures and signs in a bracket, is the name of the result of a process UNREALITY 6i which may, or may not, be actually performed. We may speak of V — i, for example, without ever performing the process of finding the root, or we may use names like quotient, sum, product, &c., though the actual amounts remain for ever un- known. As results of future processes, they are at present admittedly non-existent. A rather less simple instance of the same thing may be found in such a name as yearly income. To speak of our current yearly income in the present tense, as something actual, requires a tacit understanding that (for a passing purpose, at least) doubts as to the future continuance of it shall not be raised. There is a hypothetical element in the conception, which we recognise in theory, and yet choose in practice to leave occasionally out of sight. A still less simple instance would be such a name as the trough of a cyclone ; this is something, the existence of which depends on the relative movement of a cyclone — itself a shifting and relative state of things — and a (real or imaginary) observer. The ' trough ' is an imaginary line drawn through the lowest point of the atmospheric pressure, at right angles to the direction in which the cyclone as a whole (or else its observer) is moving. And the conception is further complicated by the fact that the forward movement is highly irregular and 62 DISTINCTION difificult to foresee, and that the cyclone itself may- more or less rapidly cease to exist ; so that even where the movement has been in one direction for,, say, the last twenty-four hours, we have no gua- rantee that at the present moment the direction is still the same ; and consequently no guarantee that we ever know where the ' trough ' is, as a present existence. And there may not be, just now, a trough at all, for the movement may have ceased ; nor, in fact, any cyclone to have a trough, for the atmospheric depression may have filled up since our last observation was taken. There is here hypothesis behind hypothesis, and when we treat the ' trough ' as something that can be spoken about, we have to shut our eyes to its highly hypothetical character. And the same is true of the ' things ' denoted by names like defect or minus quantity, where even the total from which the sub- traction is not yet actually made may never be actually in existence. Here, also, it is the hypo- thetical character of what is named that leads us to class the ' thing ' as unreal, as potential rather than actual. The question is still sometimes debated whether ' attributes ' — for example, length or hardness — are anything ' real.' We have decided ' that we will ' Page 59. UNREALITY 63 not try to answer that question as it stands. But no one, I think, denies that language, rightly or wrongly, assumes the reality of attributes, if only for the passing purpose of predicating about them. Whenever we make a name of an attribute into a substantive, so as to use it as a subject term — e.g. ' Familiarity breeds contempt ' — we speak for the moment of that attribute as if it were something real. At the same time, common-sense is quite inclined to recognise that attributes are compara- tively fleeting and dependent things, less durable, that is to say, or less substantial somehow, than the things or persons to which or to whom they belong. How far, indeed, common-sense will go in the direction of admitting the unsubstantiality of what commonly passes for ' things ' it is impos- sible to predict ; for although the habit of common- sense is to take things as real without much enquiry, yet no very deep philosophy is needed in order to see that the general distinction between things and states of things is unsatisfactory, and that our names for things are often only names for the way something else happens to affect us. A name like the horizon will serve to illustrate the latter case ; when we speak of an object being ' on the horizon,' we know very well that if we could suddenly be transported to the place where that 64 DISTINCTION object is, the ' horizon ' would be found elsewhere ; and the same of a rainbow, or the path of light made by the moon on the sea. Of the difficulty of distinguishing things from states of things, examples are very numerous : is a headache a thing, or is it only a state of the head ? Is an echo a thing ; or a flame ; or the human soui ; or the human dody ? Is snow a thing, or is it only an accidental state of water ? And is water, for that matter, anything more than an imperfectly stable condition of its two component gases ? We very soon find that, though it is possible to distinguish between any given thing (material or immaterial) and its own accidental attributes, it is impossible to distinguish between things and attributes at large ; for attributes themselves have attributes,' and in strictness even the most material things that cortie within our experience are only accidental states of something else, on which they are therefore de- pendent. Just as hardness or whiteness only exist where things are hard or white, so rocks and snow only exist where matter has taken those forms. Pure formless matter we never find, but only some perishing aspect. This or that fact of experience may be regarded either as a thing, or as ' For instance, attributes like health, strength, wisdom, &c., may have the attribute desirability. UNREALITY 65 a state of something else, according as we wish to speak of its variations, or of that nearest com- parative constant of which itself is a passing form. I here expressly avoid trying to state the case for those who quarrel with common-sense on this subject to the greatest possible extent. From our point of view they may or may not be right. It is quite possible, for instance, that ideals or ' universals ' are the most real of all realities, or that in general that which seems most, real to common-sense has least of true reality. But even supposing that is the case, the value of the dis- tinction ' real unreal ' on a lower and less meta- physical level remains. On that everyday level there seem to be two chief questions on which the decision whether a thing shall be regarded as real is generally held to turn — the question as to the durability or permanence of the ' thing,' and the question as to its independence. Both of these have to be rather broadly and loosely decided in order to use them at all, for, of course, in perfect strictness nothing within our experience is either everlasting or quite independent of other things. But of the two, the one that seems most often to be taken as a satisfactory reason for admitting reality is independence — at least in the form of individuality. 66 DISTINCTION So far as we can speak of common-sense as if it were an organ with only one opinion, I think we may say that the inmost behef of common-sense about reahty is that the real is the producible ; by which I mean only, that where the question arises whether this or that is real, we naturally ask whether a concrete instance of it can anyhow be given. Facts are the test to which we try to make appeal. That, for example, is the line that the controversial critic often takes when he wants to suggest that an assertor is talking vaguely, or using general names without enquiring what are the things to which they in fact apply. And hence the so frequent recurrence of the doubt whether attributes are real ; even the most perceptible attri- butes of things have this disadvantage as compared with the things they belong to, that they cannot be produced as pure concrete examples. If we talk about sticks or stones, the general name can have its meaning not only defined in words but illus- trated in actual cases ; on the other hand, if we talk of the length or the hardness of sticks and stones, we cannot produce an example of pure length or pure hardness, but only something which is more or less long or hard, something, therefore, in which length and shortness, or hardness and softness, are combined in certain proportions. And though we UNREALITY 67 may recognise the existence of some attributes — e.g. whiteness and blackness — which can be found apparently unmixed with their own contradictories, yet even these are not producible by themselves. The examples produced are never examples of whiteness or blackness existing independently of something which can be known through other qualities also ; an attribute is always only one aspect among many that belong to this or that producible thing. As soon, for instance, as we define ' culture ' in any sense narrow enough to apply with exactness, the actual men of culture will be found to possess other qualities, some attractive and some repulsive, in very various degrees. One way to obtain a general view of the extent of the practice of naming things which are confessedly unreal is to notice broadly the manner in which substantives are formed from words that are other than substantival.' Everyone knows that the (substantival) names of attributes — length or hardness, for instance — are mostly formed from adjectives, and that we may turn the infinitive mood of a verb into a substantive at any moment for a passing purpose — as in ' to err is human ' or ' speaking makes a ready man ' ; but the ease with which almost any word may be pressed into ' See Appendix, p, 257. 68 DISTINCTION service as a subject term, thus becoming for a time at least a true substantive — the name of something which is regarded as substantial — is perhaps not sufficiently recognised in the books on logic. Thus, nouns may be formed from adverbs, as an alibi or a tandem ; from prepositions, as a denizen {dans); from noun and preposition together, as diapason, parterre, ephemera, sinecure ; from cases of other nouns, 2& folio, specie, rebus, omnibus ; from a single word of a sentence, as creed, requiem^ quorum, dirge. Some of these things {e.g. a folio, a denizen, an omnibus) are as material as anything can be, and are apparently named in much the same way as when we form an ordinary substantive from an adjective.' The dans of denizen seems as adjectival as native or immigrant, the ablative _/^/«b as adjectival as manuscript, the dative omnibus as adjectival as a sociable or a sulky. And though a tandem is immaterial, being neither the cart nor the horses, but the way the horses are harnessed, this name is perhaps rather hard to distinguish from the adjectival name of an attribute ; and the same is true of a sinecure, or, again, of a sine qud non. But when we speak of proving an alibi, the substantive, ' alibi,' is an abbreviation for ' the fact that he was elsewhere.' An alibi can hardly be ' As to the extent of these cases, see p. 174. UNREALITY 69 regarded as itself something substantial, nor yet as an attribute. So when we speak of forming a quorum, it is difficult to say what is the nature of that which is formed ; it is rather an event than an attribute, and it is no more material than {_e.g^ a majority or an average. The name is an abbre- viation for a fact that would otherwise take a sentence to express. These instances help us to see that we should hesitate to assume oflfhand that one of the necessary functions of every name is to denote something real. It is here suggested that the notion that names are often abbreviated hypothetical sentences may help us to smooth away much of the difficulty that is traceable to that faulty assumption. Wherever the name of a quality is made into a subject term {e.g. ^familiarity breeds contempt') the meaning may be as clearly expressed in a sentence begin- ning with ' if,' or ' when,' or ' in proportion as,' or some other conditional phrase : when we say that ' familiarity breeds contempt ' we generally mean that if (or when, or in proportion as) people or things become familiar to us we lose some of the dread or reverence we felt when they were strange. Unreal distinctness, then, in our sense of the term, is only unreal for those who find it so — for 70 DISTINCTION those who in trying to interpret a name are struck with the artificial hardness of the line between that name and its contradictory, in contrast with the apparent continuity of Nature. And nowhere do I mean by the epithet ' real ' (or ' unreal ') any higher or more metaphysical kind of reality than this. The distinction is used, and is useful, even where it is known to be thus restricted ; and, in fact, until the most real reality is satisfactorily discovered I see no better way of interpreting the distinction. Till that time comes, I suppose, our views of the facts of Nature will seem to all of us more realistic than the names we use in describing them, and distinct- ness will seem artificial wherever the difference,, in nature, seems to be one of degree. CHAPTER VI IS NATURE CONTINUOUS THROUGHOUT ? If it be at all violent to say, as we said at the end of Chapter IV., that the ' laws of thought,' ' are false in every case as applied to actual things, yet it is rather a stale remark than a violent one that Nature is continuous throughout. And the former truth certainly follows from the latter. If Nature is continuous throughout, then A and non-A are always (really) one ; and however sharply they may be distinguished in idea, still there is (really) an intermediate region between them. ' The form in which the ' laws of thought ' are usually given is as follows : — Law of identity : A is A. Law of contradiction : A is not non-A. Law of excluded middle : A is either B or non-B. {i.e. S is either A or non-A.) These laws are true of the concepts A and non-A ; but, for application to actual cases, they need correction. For (i) Any actual A has been non-A and will be non-A again ; it has there- fore some non-A in it ; (2) Any actual A may deserve to be called non-A ; and (3) Between the actual A's and non-A's there is always a middle region, or borderland. 72 DISTINCTION The question as to the continuity of Nature is perhaps best treated as a question of what can be conceived. On looking closely we find that it is inconceivable ' that Nature can be other than con- tinuous. For in the first place there is the same difficulty in accepting an apparently sharp distinc- tion for really sharp as in accepting the notion of creation ex nihilo. In order that A shall be really distinct from non-A there must be no gradualness in the process of creation of A no A, unfinished, no germ of A, no raw material out of which A is pieced together. There must, therefore, be no time occupied by the process, since time is infinitely divisible ; nor must the finished A occupy any space, for a similar reason. And we cannot escape the question how any actual thing called A came into its present form of existence ; any actual thing called A must be either something permanent or something transient. But what things in Nature are really permanent when we come to close quarters with the question ? That vague ideal entity ' matter ' may be indestructible, but no actual (producible ^) form of it is so ; and in the end we seem driven to admit that the only true ' substance ' is something so indeterminate that nothing de- scriptive can be said of it. It 'exists' as the ' See Appendix, p. 259. ' See p. 66. IS. NATURE CONTINUOUS THROUGHOUT f 73 subject of change, and is only to be caught in the act of changing. Then A, if it be actual, de- scribable, producible, verifiable, must be transient, arising out of non-A and passing into it again. So that A and non-A are each of them only passing forms of the other. In this way, therefore, the picture we get of distinctions in general is that they are really fluid, but artificially hard ; that the apparent absence of a borderland between (actual) A and non-A is a result of our incomplete powers of vision wherever it is not a result of deliberately shutting our eyes to some of the facts. Either the transition is too quick for our clumsy observation, or in some way the process is hidden from us at present, and is therefore liable to become manifest whenever our observing power, or our insight into past history or remote places, shall become sufficiently improved. If experimental confirmation of this view is wanted, we have it everywhere along the line of the progress of knowledge. Every addition to our knowledge of Nature is an addition to our know- ledge of some process — an intermediate step where formerly a gap existed ; and every now and then some long-received distinction is thus found to need revision. The history of the growth of know- ledge is a history of the dissolving of older and 74 DISTINCTION harder distinctions ; a history of the discovery (or enforced clearer recognition) of borderland cases. There is no school like science — ' actual facts ' — for learning the artificiality of distinctions, and the purely human origin of words, their tentative and provisional character, and the danger of slavery to them. The practical man, however, is apt to remain unsatisfied with doctrines so ' theoretical ' as these. 'He finds himself in the presence of a ' merely metaphysical ' difficulty, to which he is ready per- haps, for the sake of peace, to bow with polite condescension, but with which he desires no nearer acquaintance. These truths are all very well on paper, he admits, but they only concern those who have leisure to play at philosophising. And he turns with relief to the well-tried doctrines of common-sense. His decision is partly moral and partly intel- lectual, and it is only the latter explanation of it with which we have here to do. On the intellectual side, what chiefly obscures the question is the diffi- culty of proving the continuity of Nature in certain cases. A notable example of the distinctions in ' which we cannot at present actually trace evolution, or gradual difference, is given by the chemical ele- ments. The chemical elements, or, let us say, most IS NATURE CONTINUOUS THROUGHOUT? 75 of them, do not alarm us or excite our hopes by threatening to shade off into each other ; that which is not gold, at any rate, never progresses a step on the way to becoming gold, though formerly the hope of its doing so guided the course of chemical experiment for ages. Doubtless it may be theoreti- cally true that actual specimens, or portions, of gold are always more or less impure ; but the range of really discoverable impurity is limited, so that this theoretic truth is of limited practical value \ to make much of it seems as nearly wasted ingenuity as anything can be. Then there are distinctions whose validity can only be theoretically denied — the geometrical dis- tinctions, for instance, like that between straight and crooked. No one in practice concerns himself with the reflection that a line as drawn is never perfectly straight, or that since it has breadth it is not a line at all. We are now speaking of the application of distinctions, and we find that in these a:nd similar cases the distinctions can be actually applied so as to satisfy the severest practical scrutiny of the most ingenious critic' ' The doubt in these cases shades off between the extremes of more and less far-fetched or unpractical. Such distinctions as that between the organism and its environment exemplify various stages of practicality in the doubt. For instance, do parasites, generally, belong to the organism or to its environment ? Some are more. 76 DISTINCTION The vagueness of the line between animal and vegetable only comes into sight when we push the enquiry to an unusual length ; and even the dis- tinction between truth and falsity can be put to practical service by every candid and sensible mind. Another, kind of distinctions might also be admitted as real in a practical sense. Those are the distinctions which are only intended to be applied within a specially restricted field, and where, accordingly, doubtful cases are practically prevented from arising. If we take a ballot-box and put only black and white balls into it, we need not fear that 'any of the balls when drawn will be grey. And a good many of the distinctions in common use are essentially of this nature. They are made on purpose, among known individual cases, rather than intended for application to unknown or future cases that may occur. We feel no doubt in applying the distinction between the classes of railway-carriages, or between towns and cities, peers and commoners, clergy and laymen, crossed and uncrossed cheques. Such distinctions are carefully guarded against difficulties of appli- others less, disadvantageous and so oppo.sed to the organism, less or more advantageous and so necessaiy to it, until it becomes hard to say where parasitism properly begins. Are blood-corpuscles para- sites? IS NATURE CONTINUOUS THROUGHOUT? 77 cation. They are made avowedly for the purpose of being applied, and the distinction-marks are made as few and as definite as possible. I do not mean, of course, that mistakes can never occur, but that where the facts are known there is in these cases no difficulty ; while in the case of any natural scale, like youth and age, or heat and cold, no amount of knowledge of the facts will remove the difficulty. Rather, it is an increased knowledge of the facts that has made the difficulty apparent. And sometimes the sharpness of the actual line is accomplished by natural circumstances instead of by legal or conventional definition. Suppose, for example, that biologists are justified in their present conception of the fluid nature of species, still certain existing species {e.g. existing men and existing apes) may very well be clearly distinct. The actual missing-links, being individuals, have died long ago, and their least-changed descendants may be changed enough to justify the sharpness of the (present) line. Again, the society of a small town may provide no examples of the truly un- decided politician, or a given district in Africa no mulattos.- It should be confessed, however, that the line between these cases and those where the field of search is unrestricted can only be justified by the reader's leave. As the field of search is widened, 78 DISTINCTION the occurrence of doubtful' cases becomes less and less improbable. In a given village, at a given time, we may . perhaps know all the voters and their political views, but increase the size of the village or town and this security vanishes ; or the line between two existing biological species may be really sharp, while if we extend the time considered, so as gradu- ally to include more and more dead individuals, doubtful cases will sooner or later occur. Still, it seems that we must admit certain cases of prac- tically real distinctness due to natural or artificial restriction of the field of search, as well as some that are due to the practical limitations of our vision or testing power. Such facts as these help us to see why the prac- tical man thinks lightly of the doctrine that Nature is continuous throughout ; or, at any rate, why he regards its purpose as fulfilled as soon as he has filed the admission and laid it away on an upper mental shelf And we may gladly follow his ex- ample, to the extent at least of fixing attention chiefly on the definitions that are troublesome in practice. Our present purpose is to show the ex- tent and importance of unreal distinctness — that is to say, of the disagreement between definite lan- guage and fluid facts, — and I only mention the seemingly doubtful examples in order not to rest a IS NATURE CONTINUOUS THROUGHOUT? 79 strong case on weak supports. For I hope to show that, in spite of all the apparently real distinctness that exists, there is also sufficient adnaittedly unreal distinctness to warrant our raising the question how far it is to be regarded as one of the permanent sources of faulty thinking and of needless heat of controversy. CHAPTER VII COMMON-SENSE AS DIVIDED AGAINST ITSELF It is difficult, as everyone knows, to discriminate between paradoxes ' which are interesting and those which are flat and trivial. This is partly because of the fact that one person's interest in a given question is different from another's, and that each person's interest in them varies from time to time, but also because the difference is really a gradual one. This or that paradox may deserve any shade of treatment between the utmost respect and the utmost disrespect. And the question occurs to one. Can any general account be given of the qualities or conditions that make a paradox respectable ? How is it that some are only foolish or dangerous playthings, while others give us a vision of un- familiar truth ? We are here concerned specially with only one group of paradoxes, but what is true of them in ' ' Paradox ' is here used in its older sense, as simply ' a de- parture from received opinion.' COMMON-SENSE AS DIVIDED 8i this respect is true of all. We are concerned with the judgment that a distinction which claims to be sharp and commonly passes for sharp is, strictly speaking, rough, or artificially sharpened. And we may freely admit the existence of cases where this judgment, though literally correct, has little or no practical value. But it is worth remarking that in all cases what chiefly spoils the value of these or ■of other paradoxes is the quality of staleness. It is not necessary that a doctrine should be perfectly true in order that it shall possess interest and value — perfect truth is so difficult of attainment that even the wisest human minds find interest and value in truths that are imperfect ; but it is necessary that a doctrine, besides containing a certain amount of truth, should somehow avoid the appearance of staleness — of having been seen, ad- mitted, and pushed aside as trivial. For the charge of being already admitted is, when justly brought, the most damning charge that can ever be brought against an assertion. At least, the truths that are worth asserting are never those that our audience fully accept as true. And staleness, of course, is a quality that para- doxical opinions are apt to possess unawares. There are many kinds of paradox that attract us most when we are young and inexperienced, or G 82 DISTINCTION that attract especially those who know little about what others have thought before them. And since it is easy to call a truth stale as soon as we see that we really must admit it, this is perhaps the charge most frequently brought by common-sense against,, for instance, the doubts about drawing the line in Nature. These flux-theories, we are told, have been before the world since the time of Heraclitus. Everyone knows all about the difficulty ; no one in practice minds it now ; if we must have a puzzle to play with, let it be something rather less old and worn. In order to give due weight to this natural form of objection, I have referred expressly above ^ to the part played by ' common-sense tact ' in the use of distinctions. And, indeed, we cannot, I think, pretend to overlook the fact that common- sense is certainly not always deceived by unreal distinctions — that it has to a great extent acquired the art of taking and using them lightly, with knowledge of their faults. And so it must often happen that the laboured proof that a distinc- tion is faulty wears a perfectly genuine look of staleness ; we never really forget, for instance^ that a continent, like an island, is ' land surrounded by water,' so that only a gradual difference of size ' Pages 17, 36. COMMON-SENSE AS DIVIDED 83 is referred to by the distinction ; or that a child becomes a man by imperceptible stages ; and in plenty of similar cases we are fully accustomed to take the distinctions ' for what they are worth/ and to use them gently. And if we could somehow distinguish these cases from those where common- sense is actually deluded, we should then have dis- covered what are the more important kinds of unreal distinctness. Few things are more difficult than to prove common-sense in a given case more rigid and clumsy than it claims to be. Tact in the use of distinctions is, of course, not possessed by all men equally, nor by any man equally on all occasions, and even our most effective tact of to-day will doubtless be later seen as insufficient. But these are merely general considerations, and the special case must always stand on its merits. How, then, shall we ever distinguish safely between the false and the justified claim to have duly discounted a given distinction ">. Fortunately, we can separate to some extent the disputed from the undisputed claims. The cases where common-sense tact is even apparently successful certainly do not cover all the ground. We find a considerable number of questions in 84 DISTINCTION regard to which common-sense opinion is strongly divided as to the stress that ought to be laid on a given distinction ; we find, that is, more than one grade of self-styled common-sense, or insight into realities. Philosophy, religion, worldly wisdom, science, poetry, each in the name of insight over- ride distinctions that the others are wont to mag- nify, or dwell on distinctions that the others regard as misleading. At any period in the history of thought there is conflict of opinion as regards a large number of stated doctrines ; and, as noticed above,' in all the more lasting conflicts of opinion some difference of view as to the manner in which a name is to be interpreted, a distinction applied, lies at the root of the controversy. That, then, appears to be a sufficient answer to the objection that we have heard all about the continuity of Nature long ago. We have heard many things long ago that we need to hear again. It is only quite lately — only since 1859 — that common-sense has begun to make much use of the notion that distinctions are fluid, and the two stages of objection that come before the claim to ' have known it all along ' are hardly yet left behind. And though tact in dealing with fluid distinctions is everywhere to be seen at work, some failure in ' Page SS. COMMON-SENSE AS DIVIDED 85 its working is shown wherever common-sense is divided against itself. And, broadly speaking, we find that it is im- material things that raise most disputation. In regard to material things common-sense is perhaps seldom much at fault. The most treacherous kind of unreal distinctness occurs where the ' thing ' that is named is (like culture) never foynd pure in the concrete world at all. It was suggested above that truth and falsity are possibly things of this description. To find a satisfactory example of a perfectly true statement, or even a perfectly false one, is held by some philosophers to be a task of infinite difficulty. Animals and vegetables, islands and gold and children, such things as these are easy to find in Nature ; they do not require to be imagined by an effort, or sifted ideally out of a larger whole, from which they are never found separate. They are what common-sense everywhere agrees to regard as things. But many things that are named are admittedly things by courtesy only ; common- sense is in two minds about their real reality, but is willing, nevertheless, to regard them as real for the moment. We need not here press the example of truth and falsity, if by chance it appears far-fetched. There are cases enough, as we noticed above,' ' Chap. V. pp. 60-69. 86 DISTINCTION where common-sense is fully aware of this kind of unreal reality, and plenty of other cases where it is perfectly able at times to become aware of it. These immaterial entities that so frequently lead common-sense into an actual difficulty are sometimes called ' abstractions.' An abstraction, as such, is something simpler, purer, more ideal (whether better or not) than what we find in Nature. Evil, for instance, is an abstraction — actual things being more or less evil, but never entirely so ; if required to give an example of unmixed evil, among phenomena, we could not find a perfectly unex- ceptionable case. And the typical abstractions are what are commonly called qualities, or attri- butes. Attributes, when named in order to be spoken about, are always something abstract ; as we noticed above, they are not found existing in- dependently of things that possess them in greater or less degree. Thus, when we analyse some total into parts that are not found separate from the whole — e.g. the various ' faculties ' of the mind, memory, imagination, and the like, or the various aspects of anything, or the list of qualities essential to any species — the parts distinguished are abstrac- tions ; they are not producible purfe. And the reason why abstractions are so troublesome is that the roughness of a distinction between producible CO.VMON-SENSE AS DIVIDED 87 things — things of which a concrete instance can be given — is comparatively harmless, since the correc- tion of these names by facts is so much more easily managed. In recent times, at any rate, the habit ■of remembering that concrete things are for the most part named and classified somewhat roughly has taken root even in popular modes of thought, and is growing and flourishing strongly before the eyes of the present generation ; so that it has become an anachronism to let the names of natural objects greatly obscure the facts about them. When natural objects are in question we go to Nature at first hand in cases of diiificulty, instead of appealing to the guesses at fact that are handed down in language. CHAPTER VIII SPOILT WORDS It appears, then, that the custom of naming un- realities as if they were real is the chief source of unreal distinctness of the practically troublesome kind. But I wish, further, to show in a more com- plete way the extent of the influence of unreal distinctness upon ordinary modes of thought. For the reasons mentioned in the preceding chapter we shall say little or nothing of its possible influence in the cases where substantial agreement is reached by the mass of opinion which can fairly be called orthodoxy in any department of knowledge. Though I confess to a pious opinion that even there unreal distinctness is operative — in fact, if there be such a thing as progress in knowledge, this must ' be so — yet the claim that common-sense makes to be able to treat rough distinctions lightly and ' ' Must : ' for all incompleteness of knowledge expresses itself in unreal distinctness (or ' abstractness ') ; which in turn acts as a drag upon progress. See pp. 73, 99, 242-245, 248. SPOILT WORDS 89 safely seems to me too plausible to be disproved in given cases, except where common-sense is plainly- divided against itself. But this is a large exception. It implies that unreal distinctness is at work wherever one man's opinion conflicts with another's, or wherever our judgment is pulled two ways — so far at least as the difference of opinion is irreducible by an appeal to ' facts.' And the range of what may be called ' ideal disputes ' becomes a great deal wider when considered in its beginnings than when we connect the notion, as we did in Chapter IV., with only the more striking developments of it. These great growths of the party spirit have had their origin in little germs that are hardly noticed at first. Words pass current not because they mean the same to everybody, but rather because we can manage to agree to keep the differences out of sight ; if we once begin to be universally strict about definition we miss not only the charm of ' fluid and passing ' intercourse and ' the general graces of conversation, but something also of much greater importance philosophically — namely, the habit of distinguishing between more and less dangerous vagueness, and of seizing on the broader outlines of meaning which are needed to lay the founda- tion of any efficient attention to detail. So 90 DISTINCTION that forces are always at work to smother con- flicts of opinion, which nevertheless are only- waiting for a breath of wind to fan them into a blaze. As illustrative of this stage of repressed ideal conflict, I would choose the class of what may be called spoilt words. By these are here meant not only cant words ' — expressions used because they are somehow fashionable — but also and specially all words whose sense has been vulgarised by their being applied to the more showy, more easily under- stood forms of that which they try to name, as, for instance, the words genteel or worthy have suffered in their time,^ or as the word clever is suffering to-day ; till at last to call anyone ' 'genteel ' would be, in effect, to call him ' ill-bred ' ; to call anyone ' worthy ' means that we think lightly of his worth ; and soon, perhaps, we shall only call people clever when we mean that they are narrow and sharp and ' And there are other forms of the spoiling of words, which do not concern us here. For instance, the case where an often-used phrase influences the meaning of one of the words composing it. Thus the word incarnate, from its frequent connection with fiend, appears to mean to some people almost the same as infernal. ^ It should be noticed that spoilt words do not remain ' spoilt ' for ever. After a time they simply take a new (unambiguous) meaning. If the reader thinks the words ' genteel ' and 'worthy' have not yet done so, perhaps the word silly, which once meant nearly the same as hdppy or blest, or the modern German schlecht, which once meant nearly the same as good, will serve as examples. SPOILT WORDS conceited. Spoilt words, in short, are all words that perform their descriptive function clumsily, by failing to tell us which of two very different things is meant ; words which are used by wise and foolish alike, but with different application. The words wise sxid foolish are themselves examples in point, and in fact all common and current epithets which impute praise or blame or contempt are extremely liable to grow into this condition. The child condemns or praises with a broad epithet where the man can find no suitable word. The sentimental and the cynical youth condemn ' cynicism ' and ' sentiment ' in a more sweeping manner than they will care to condemn them a few years hence; and the very people whom I to-day should describe by the uncomplimentary titles just used, may be seen by my wiser neighbours to be ' good hearted ' and 'sensible ' respectively. As the Master of Ballantrae remarks, ' there are double words for everything : the word that swells, the word that belittles. . . Call it vanity, call it greatness of soul, what signifies the expression ? ' When we find vague words like these applied by a person whom we think less knowing than ourselves, we feel that some protest or objection is needed if we are to guard against mistakes of meaning ; the spoilt word seems clumsy to us. DISTINCTION flat, deficient in point ; it blurs impoi'tant differ- ences which are of the essence of the assertion supposed to be made. Our claim to fuller know- ledge of realities {i.e. of actual cases) leads us to distinguish where the other person sees no differ- ence of meaning ; we have learnt to recognise various sorts of cleverness, courage, &c., and accordingly find the bare name ambiguous ; the epithet clever, for instance, does not tell us what kind of mental excellence to look for — breadth, or depth, or tenacity, or merely quickness, or possibly cunning ; the epithet brave leaves us uncertain how much of the better part of valour to expect, or how far the bravery requires an audience or depends upon coarseness of fibre, or dull percep- tion ; the epithet generous may merely mean im- pulsive and short-sighted, or fond of popularity, or reckless of expense. And the chief difference between intercourse with our equals and with those whom we fancy to be less experienced is that in the latter case so many useful notions are rendered useless ; a Christian finds the word faith unsatis- factory as used by an unbeliever ; a philosopher finds the word /ro^T/" applied to all sorts of faulty evidence ; and the specialist everywhere feels the same sort of check upon intercourse with outsiders that is felt by us all in speaking in our own SPOILT WORDS 93 language to a foreigner possessing only a smatter- ing of it. The finer hints of meaning are absent from between the lines of the talk ; words have to be chosen for their simple harmlessness rather than for any positive kind of excellence, and the style and the matter alike become cramped and meagre. But the spoiling of words is, from the nature of the case, a delicate and unpopular subject to treat in any detail, just because there is so little agree- ment as to which words are spoilt and which are not. For instance, I cannot hope that even the few common examples given above will satisfy every reader. To a certain extent the whole tend- ency to claim that given words are spoilt is dis- liked by common-sense, as being an outcome of the exclusive or priggish spirit, or at best pedantic and tiresome. Yet, fortunately, here again we find common-sense at various levels, and so divided against itself ; even to the commonest sense some words seem vulgarised, the main difference in this respect between the higher and the lower grades of realistic insight being that every higher grade is inclined to find more words thus faulty than the grades that lie beneath it. The commoner the sense, the more ready it is to accept as suflficient the rough, uncriticised notions that come first to its 94 DISTINCTION hand. It is rather the literary critic, for instance, than the general reader, who finds any fault with the distinction between the novel of incident and the novel of character, or between realism and idealism in art. The best apology that we can make, however, for seeming here to deride all popular modes of thought is to admit freely that, if the use of clumsy descriptive names were wholly unjustifiable, the class of perplexities thence arising would be far less troublesome than they are. Common-sense, as usual, is here in close sympathy with the genius of language. The ' general name' itself — the basis of language — exists by neglect of minor differences in favour of broad resemblance. One of its func- tions is to give us an outline which riper experience may modify. Hair-splitting is only possible when the hairs are already there to split, and if our notions are to become refined and more realistic, the simpler and more wordy notions must somehow be gathered first. Besides, it is hard to conceive the total disuse of our first rough sketches of things. Not only have they a value for children, or for those who live in the hurry and press of business, but their necessity comes home at times to everyone. If Nature is continuous, all general names are to some extent liable to the fault we are SPOILT WORDS 95 here discussing. The insufficiency of descriptive words to perform their descriptive function is never entirely conquered — that is to say, the individual case is always richer in detail than any or all of its class-names strictly indicate. The clumsiness of descriptive names is, therefore, itself only a matter of degree and occasion, and the permanent problem is not the impossible one of avoiding all such clum- siness, but the practical one of avoiding just so much of it as is relevant to the matter that happens to be in hand. We need rough names, then, as well as exact ones. Just in the same way as all general names are useful — that is to say, in enabling us to blaze a path through a forest of troublesome details — so the names which are most open to the charge of clum- siness are for that very reason most useful in their place. If we try to do without any one of the words just quoted as examples, we are conscious of a loss ; it is a pity, we feel, that so useful a name as clever or wise or generous should be thrown away altogether as condemned by its wrong applications ; and, in fact, where a meaninghas become completely reversed, we generally do invent some other way * of expressing the original meaning. Though ' Or else we show, by our hesitation in choosing a word at all, our regret that all the available words are spoilt. 96 DISTINCTION the word be clumsy on certain occasions — for instance, in a testimonial written by a stranger — yet life is not long enough to give us a chance of always refining our notions to the utmost ; there are times when the slight ambiguity causes less harm than would come from the check to our thoughts. And hence it is that this source of trouble belongs to the permanent class. In the use of language we are beset by two opposite dangers — blindness to important differences, and confusion or waste of time through attending to unimportant ones ; and the practical problem is always to find the proper compromise, or rather to distinguish with increasingly greater correctness the relative importance of differences in regard to a constantly shifting and various set of occasions. The chief purpose of Chapters IV. to VIII. in- clusive has been to show the extent of unreal distinctness that exists in language and in our habitual ways of thought. We have found that in perfect strictness every ' general notion ' suffers from this fault, and that even if we are content with a more practical view — content to notice only so much of it as, through being insufficiently dis- counted, leads to confusion of thought as shown in difference of -opinion — the extent of its influence SPOILT WORDS 97 is considerable. All words that perform their de- scriptive function clumsily are, in so far, ambigu- ous, and the clumsiness of descriptive words is relative to the completeness of description desired. So that wherever two minds desire completeness of description unequally, the words used in de- scribing will appear more clumsy to one than to the other. And as a matter of fact we find that the effects of insufificient description begin to be widely felt as soon as we talk about immaterial things. Where the faults of a name can easily be corrected by reference to facts, words do not — in modern times at least — greatly mislead us. But abstrac- tions or ideals are likely long to remain a source of bewilderment and error. For their names are always of loose application, and mean different things to minds at different levels of realistic knowledge. We have next to enquire more pre- cisely what effects may be produced by the use of ambiguous words. H CHAPTER IX THE EFFECTS OF AMBIGUITY Ambiguity, or clumsiness of description, has two very different sets of effects, the difference depend- ing on whether the clumsiness is recognised. Unseen ambiguity shows itself in what is usually called narrowness or hardness of view. This fault has been already mentioned more than once — for instance, in speaking of ' idealisation and caricature.' We noticed there how the difference between op- ponents is exaggerated, and how the truth which is confusedly seen by each is hidden from the other, by the artificiality of much of the opposition be- tween them. When the line between Conservative and Liberal, for instance, is drawn too sharply, actual mistakes of fact are made as to the real aims of each of the opposite parties. But the direct effect of unsuspected roughness in a dis- ' Chap. IV. THE EFFECTS OF AMBIGUITY 99 tinction may be understood better by reference to our remarks on the spoiling of words. The am- biguous word and the clumsy word are the same. To believe too rigidly in distinctions is the same as to generalise superficially, or to believe in a rule without understanding its conditions and limita- tions, and so without making allowance for the exceptional cases. So that unseen ambiguity is the same as excessive abstractness or thinness of view — insufficient complexity and attention to detail — a fault which, so far as knowledge is pro- gressive, is gradually being conquered. But in the act of seeing the unreality of a dis- tinction we lose sight of the meaning which that distinction gives. One way of describing the pur- pose of distinctions generally, is to say that they exist in order to give predicative meaning to names. That is not the explanation of how every dis- tinction arose — for the distinction between light and darkness, for instance, is probably drawn by animals that are far from having a language — but wherever a name exists with a meaning, there a distinction is drawn, inevitably, by the mind that apprehends that meaning ; and on the clearness of the distinction, to that mind, the definiteness of the meaning depends. By ' predicative meaning ' is meant what J. S. Mill and others have meant by H 2 DISTINCTION ' connotation.' ' For convenience we will here call it ' meaning ' simply. The broad fact that the meaning of names everywhere implies and depends upon distinction is not very difficult to grasp. Of language wholly free from distinction, indeed, we have no experience and can form no steady conception ; we cannot observe, or even imagine except quite negatively> what would become of a language if all its dis- tinctions were swept away ; but we can easily observe the effect of the disappearance of this or that distinction. And though we may not be able to find a case of the total or permanent disappearance of a distinction that has once been seen, yet partial or temporary disappearance — partial or temporary identification of A with non-A — is common enough. This occurs, in fact, wherever we recognise that a distinction strictly ' The . ' connotation ' of a name may be explained as the con- ditions under which that name is intended to be applicable — applic- able as a predicate to any subject. If a name A has any predicative meaning, no subject (S) can rightly be called A unless certain conditions are satisfied ; A implies (or indicates, or signifies) C, for instance ; so that if S is not C it does not deserve to be called A. In this way there are always assumed to be certain facts which, if true, would deprive S of the right to be called A. For if there were no such possible facts, there would be no point in the predica- tion. Any other predicate (e.g. non-A instead of A) would do as well. By the ' denotation ' of a name, on the other hand, is meant the things or cases to which it is intended to be applied. THE EFFECTS OF AMBIGUITY loi depends on a gradual difference — wherever we think we see the continuity of Nature. And the effect of such recognition, while it lasts, is to destroy the force, the value or meaning, of each of the names A and non-A. What, for instance, is the point of the epithet sane (or insane) where the doubtfulness of the line between them is brought before our notice? The reason is, that the meaning of any word is essentially a case of ' value,' or comparison — of standing out clearly against a background. Where the outline is blurred the meaning is lost. This may be seen either by imagining an extreme case or by taking any one of the actual cases of partial blurring that are so common. The extreme case could only be furnished by a name which should profess to be applicable to everything indiscrimin- ately. Perhaps the nearest approach we have to such a name is the word ' thing ' — at any rate in such forms as 'something' 'anything,' 'nameable thing.' True, that if we ask whether (k^.) virtue is a 'thing' we can put a meaning into the question ; but this is precisely by contrasting the notion of ' thing ' with some other notion, such as < attribute,' which contrast, or distinction, serves to give point to the question. On the other hand, we should iind it impossible to put a meaning into I02 DISTINCTION the question whether virtue is ' something,' or is a ' nameable thing ' — since if it is not something there is nothing else for it to be ; and besides, in naming it we have already answered the question whether it is nameable. It is easy to see that an epithet or a class-name which might be applied to anything indifferently would be received with complete indifference if once that fact were clear — to be without the deco- ration would be more distingui. Take, for instance, a word like 'natural.' The whole point of the epithet natural lies in the existence of things that in some sense or other may be called unnatural \ were everything in the universe natural, then to apply that predicate to any particular thing would be a waste of breath. But just because language has grown up under the pressure of practical needs, no existing epithet exactly illustrates so total a want of ' point,' or predicative meaning. Indeed, the difficulty is rather to find, when we want them, words that are wide enough to deal, as some philosophers have attempted to do, with all-inclu- sive subjects like Being in general, or Nature in the widest sense. Thus, Hegel finds that pure Being is the same as Nothing, and Spinoza has to distin- guish natura naturans from natura naturata. Yet partial failure in ' point ' is common enough. In THE EFFECTS OF AMBIGUITY 103 order to found the complaint, for instance, that the predicate ' natural ' lacks point, it is not necessary to raise the rather far-fetched objection that every- thing is natural, and therefore that nothing is speci- ally marked out by the designation. It is enough that the word natural lacks clearness of definition. A natural manner, for example, is so well known to be charming, that to be natural without at the same time knowing it is a rather uncommon ac- complishment in grown-up people ; and of course to be consciously natural, and still more to be natural of set design, is to play tricks with the line between natural and artificial. We hardly know how to describe a natural manner which is just a little conscious of itself. The epithet loses its value when the line begins to melt away. So again, unless we draw a distinct lihe between infancy and man- hood the assertion that So-and-so is (or is not) of full age becomes only capable of carrying a broad and rough meaning ; and, in certain contexts at least, a rough meaning is the same as no meaning at all, inasmuch as it cannot be used for drawing consequences. If, for instance, the fact of infancy is to be pleaded as a reason for not paying debts, there must be a clear understanding as to the date when ' infancy ' ends. In this way, then, dis- tinction is required in order that any word may I04 DISTINCTION have consequence, or meaning. Distinction stiffens language, gives it shape and consistency, acts as backbone ; and where no sharp distinction exists in Nature the needs of practice often compel us to form one artificially. It is because predication always involves a choice between yes and no, that its meaning depends on sharp distinction. Assent and denial must be distinctly different in meaning if either of them is to have a meaning at all. The sine qua non of meaning in any predication is that the predicates A and non-A shall not be applic- able with indifference — as happens where the line between them is dissolved. It is this considera- tion chiefly, as we shall see in the next chapter, that makes the Socratic method of demand- ing definitions so formidable a controversial weapon. But the chief difficulty of the question lies in wait for us beyond this point. If Nature is con- tinuous, all distinctions are rough, all predications therefore incompletely definite. But surely not all predications are quite devoid of meaning ? How far, then, are we prepared to press the doctrine that meaning depends on. sharp distinction ? Is the value of distinctions itself a matter of degree, depending perhaps upon the extent of borderland, — so that names may be regarded as less and more THE EFFECTS OF AMBIGUITY 105 ambiguous and their meaning as greater and less in amount — or does meaning stand and fall with absolutely sharp distinction, real or pretended ? Let us see what is said in favour of each of these two conflicting views. On the one side it is freely admitted that the applicability of a given distinction to actual cases is an important element of its value or meaning, and that if we were to confess in a given case that the things ' distinguished ' were perfectly indistin- guishable in practice, the value of the distinction would be wholly gone. But then, we are reminded, the things that language distinguishes are never, in practice, perfectly indistinguishable ; there are always well-marked cases at the opposite ends of the scale, and the range of these nearly always extends a long way towards the middle region ; up to fifteen or later, one remains a boy ; after twenty- five or earlier, one has already become a man. Quite commonly we admit the difficulty of draw- ing a given line, and then proceed to use the dis- tinction as if it were perfectly firm and clear. This is, indeed, a leading fact about the manner in which we are all in the habit of dealing with loose dis- tinctions ; their partial failure in applicability does not, though recognised, entirely destroy their value. The distinction between right and wrong, for io6 DISTINCTION instance, or true ^.nd false, or attractive and repul- sive, does not as a matter of fact become 'entirely meaningless as soon as we admit that its applica- tion is not perfectly clear. The extent of the doubt- ful margin or borderland varies a little in various cases — that between wise and unwise, for instance, is larger than that between straight and crooked — now and then becoming perhaps slightly trouble- some ; but as a rule the extent of the borderland is as nothing, compared with that of the well-marked cases, and to pay much attention with the bare fact of its existence is a waste of time.' At any rate, rightly or wrongly, people are often content to use words with some vagueness, and are impatient with any attempts to make them definite or to enquire very closely into the meaning of what is said. Pedantry, quibbling, casuistry, and other uncomplimentary ' The notice that common-sense will accord to a given border- land appears to depend on various accidental conditions ; in some cases the difficulty is freely recognised and deplored ; in others the theoretical dii£culty is adifiitted, but its practical importance is denied — e.g. it is useful to speak of islands and contiiients though Australia is insular and continental at once ; and no one thinks it worth while to insist on the difficulty of finding the two ends of a river. In other cases, again, we are rather afraid to make the admission, or think it well for certain purposes, or certain people, to put the borderland out of sight. For example, a reformed drunkard may wisely decline altogether to speculate as to the line between moderate and immoderate drinking ; and the need for action imposes on everybody the duty of suppressing even well- founded casuistry on occasion. THE EFFECTS OF AMBIGUITY 107 names, are commonly given to the practice of trying to force a speaker to define his words ; and the claim is widely made that, in certain subjects at least — poetry and religion, for example — the aim at definiteness may very well be worse than merely pedantic and trivial. Meaning depends, according to this view, not on absolutely sharp distinction, but on the smallness or general unimportance of the borderland cases ; some words suffer much, but most words hardly at all, from defect of mean- ing through lack of clear definition. On the other side, it is freely admitted that in a certain sense we can speak of the varying extent of borderland, and that even where the borderland is a large one there are often sound practical rea- sons for determining not to see it. But then, we are reminded, in doing so we are shutting our eyes, however wisely, to a piece of admitted truth ; and the consequences of this neglected truth are there to be drawn, and cannot in fact be avoided, even if we personally prefer to hide our heads in the sand. And after all, these stricter consequences may be admitted, and then forbidden to interfere with practice. It is open to us, when once we know what they are, to make as much or as little use of them as we choose ; but lest there be any mistake as to what they actually are, it seems io8 DISTINCTION worth while just once to set them down as clearly as we can. Perhaps the most useful analogy in the light of which to see the strict dependence of meaning on sharp distinction is one that was used by Mill in an early chapter of his Logic — the story in the ' Arabian Nights ' of the chalk-mark on the door- posts. If you mark one doorpost so as to signify ' This is the house,' the meaning of the mark is clear, but if only a single other doorpost be marked in the same way the sign becomes ambiguous. The mark is then false in a single case and there- fore fallible altogether ; it may mislead us, and the most misleading sign can do no more. This analogy provides room also for the first, and more common-sense, of the two views given above ; for common-sense would certainly call the mark still more ambiguous if instead of only one house fifty or a hundred others were chalked. What corresponds to the mark is the name — for example, ' A.' What corresponds to the meaning of the mark is the name's connotation — for example, ' C Common-sense insists that where only a com- paratively small proportion of things named A are in fact not C, the name A is less ambiguous than where the proportion is larger ; and that, some- where on the scale, ' less ' ambiguity shades off into THE EFFECTS OF AMBIGUITY 109 ' practically none ' ; so that, for instance, we can use words like straight and crooked without any harm at all, and even words like wise and foolish with meaning enough to satisfy everyone who is not a mere obstructive in practical matters. The ' unpractical ' view, on the other hand, is that in strictness the gap between certainty and uncer- tainty is wider than between an even chance and any amount of odds. Though probability admits of degrees, possibility does not ; so that, once admit the smallest uncertainty in drawing the line, and any actual thing or case called A may really deserve rather to be called non-A. A word that is ever so slightly ambiguous may deceive an audience, and what worse harm can be done by the most equivocal word in the language ? Or, to put the matter in another way, unless the assertion of the predicate A carries with it in all cases, without exception, the denial of the pre- dicate non-A, how are we to know whether it does so in the particular case that comes before us ? The particular case in question may be the one excep- tion out of millions, unless we have some reason (other than the mere number of the non-excep- tional cases) to suppose the contrary. If mere numerical probability could make us certain, then it is certain that no particular railway accident will DISTINCTION ever occur, no lottery-ticket ever win a prize, and no particular bullet fired in battle ever hit a parti- cular soldier among the enemy. Perhaps these are ' practical certainties ' ? And of course if practical certainty means a kind of certainty that is daily proved mistaken in practice, no more can be said, except that such mistakes of judgment are exactly the fault we were asking how to avoid. Unpractical as this stricter view may appear, and liable, as it doubtless is, to become really un- practical when wrongly applied, it contains at least some truth that is worthy of notice. It is clear that the names A and non-A must not be regarded as indistinct, if they are to have a meaning. So long as the line is supposed to be vague, our con- fidence in using either of the contrasted names as a predicate halts in every case until some special justification is provided. Otherwise there is a leak, however small and unnoticed, which may bring down the meaning of our assertion like a pricked balloon. If the distinction is thought of as at all inapplicable — if it be supposed that any cases belong to a doubtful borderland — how do we know that the case in question is not one of them } The only general grounds, as distinguished from special grounds, for this security are to be found in the assumption that no borderland cases exist. In THE EFFECTS OF AMBIGUITY iii the absence of this assumption, and also of special reasons for excepting the given case, our confidence is seen to be pretentious and unsound. We must, therefore, somehow agree with our audience either that the distinction is strictly applicable, or else that the case in question is not within the doubtful margin. Our plan of reconciling the practical and the theoretical views will best be understood in the light of later chapters, but already we may see the use of recognising the fact that meaning depends on absolutely sharp distinction. It is not a merely unpractical and obstructive truth, deserving ' to be forgotten altogether as soon as it has once been grudgingly admitted. It helps us to correct the natural error of supposing that the smallness of extent of borderland is what makes a meaning clear, since it helps us to see that extent of border- land does not by itself settle the matter, but is only one among other conditions tending to its settle- ment in a given case. The borderland between sinned against and sinning is a very small one — the notions are not very liable to become ambiguous — and yet sometimes their meaning is lost through overlapping ; the borderland between good and evil is a very large one, and yet these words are often clearly understood. Extent of borderland is thus DISTINCTION not by itself the decisive cause of loss of meaning, but only of liability to such loss. Where the borderland is large, ambiguity is of course more likely than where it is small, but where the appli- cation of a word is actually doubtful the doubt cannot be removed by reference to the fewness of the doubtful cases ; and on the other hand, wherever an ' ambiguity ' does not matter, there no real ambiguity exists. In this sense, therefore, it may be said that meaning does stand or fall with absolutely sharp distinction. It does not require that Nature shall be discontinuous, but only that its continuity shall be by agreement forgotten at times. Meaning depends on sharpness of distinc- tion, but pretended sharpness will serve the purpose of making a meaning just as well as if the sharpness were ' real ' in any higher sense. Meaning requires agreement, but agreement does not depend upon knowledge of fact. CHAPTER X THE DEMAND FOR STRICT DEFINITION Criticism of the soundness of a judgment is so nearly allied to the accusation of vagueness or ambiguity, that it may always be expressed in the form of a demand for a definition, and is often most effectively so expressed. The cases where it is least easy, or least natural, for criticism to take this form are where there is least doubt as to how the facts should be named, and since in a great many cases such doubt hardly exists at all, or, at any rate, is of too cautious and ' theoretical ' a kind to be noticed by practical men, there is considerable difficulty in showing, in any easy and yet convinc- ing way, the complete extent of the truth that all criticism of judgment is criticism of distinction.' But we may here be content with a less ex- tensive view. Our special object, in this and the preceding chapter, is to discover exactly the nature of the harm that is actually caused by amlsiguity. ' See p. 193. I 114 DISTINCTION Our attention is thus directed away from the cases where there is least doubt as to the applicability of a name, and attracted towards the cases where such doubt becomes really and practically troublesome. These, as already said, are most easily found in controversy. And since the disputes that turn upon fact ' are short-lived and uncontroversial when compared with those which turn upon theory (or ideas, or ways of dressing up fact in language), it follows that the more we fix attention on con- troversy, and especially on the less easily settled matters of controversy, the better chance we shall have of seeing the actual manner in which ambi- guity helps to obscure our judgment. Sophists and rhetoricians early discovered that whoever puts, forward any assertion lays himself thereby open to troublesome questions as to the exact meaning of his words. In the time of Socrates this controversial weapon was, no doubt, wielded more freely and eagerly than is the custom to-day among those who debate with telling effect ; yet, less simple in manner, and restricted to fewer and rather more carefully chosen occasions, the ' Socratic method ' is in truth as powerful now as it ever was. In essence, this plan of attack consists ' So far as ' fact ' can be distinguished at all from ' theory,' See p. 1 54- THE DEMAND FOR STRICT DEFINITION 115 in putting innocent-looking questions to the assertor, and so leading him step by step into a tangle of meanings and of inconsistent assertions. You begin by humbly sitting at his feet, and asking for instruction, and you end by driving him into such a position that, if he cares even to save appearances, he must withdraw his claim to possess the requisite knowledge. ' The strength of this method depends partly upon the real difficulties that so often lie in the way of getting our mean- ings clear and our proofs convincing, but also partly upon the extreme plausibility of the humble demand for enlightenment. Since assertion as such professes to be instructive, the assertor can hardly be surprised if the audience expect him to know what he means himself Otherwise there is an air of pretence about his claim to teach us, and pretence unmasked is fatal to the prestige of a teacher. Nor can he escape even by silence very successfully. To assert and yet to refuse to explain his meaning, to impart an oracular message and there stop short, is a form of an- nouncement that, however solemnly, beautifully, or lightly put forward, can only be imposing in face of a timorous kind of criticism. The more careful the investigation, or the more the audience is really interested in discerning in what respects m6 distinction the assertion is true and false, the closer must naturally be the enquiry into meaning, and hence to press for exact definition is nearly always an easy and telling controversial trick,' full of destruc- tive insinuations, which may be made with an air of the utmost innocence, as arising from the honest wish to be enlightened. A general name which on enquiry means nothing particular, or a general assertion which on enquiry turns out to have no precise application to particular cases, is easily made to look foolish. And doubts as to precise application do not depend for their strength upon the number of doubtful cases that can be found, but upon their nature. Any one such case stands for a class of cases, just in the same way as a single figure illustrates a general proposition in Euclid. ' A trick, but not necessarily unfair ; useful certainly as against over-pretentious assertion in all its forms, and specially against the ' oracular ' attempt to throw upon the audience the burden of reconciling contradictory statements. The opposite dpctrine, en- forced so pertinaciously by Socrates, was that the assertor and not his audience — he who pretends to instruct and not he who wishes to be instructed— is the proper person to effect whatever reconciliation is possible ; and that the higher the assertor's authority, or the bolder his profession of knowledge, the heavier lies on him the obligation to make his meaning clear. Socrates felt that where an assertor put forward two inconsistent statements he may be rightly called upon to choose one horn of the dilemma. The utmost freedom of choice may be given to him, but until he chooses one or the other definitely the audience iave no real assertion before them, and he might as well have left the words unsaid. THE DEMAND FOR STRICT DEFINITION 117 Even a single doubtful case may thus upset an assertor's pretensions. He talks about justice, let us say, or calls an action unjust ; and, in order to discover exactly his reasons for applying the term, we ask him how he would apply it in some other case where (it seems to us) justice and injustice are subtly interwoven. We find that there, at any rate, he feels the same difficulty as ourselves. The distinguishing marks of justice, therefore, as the assertor himself conceives them, are hazy on occa- sion ; and we naturally want to know what is the real extent of these occasions. For all we can say, and for all the assertor has yet been able to show us, they may include the case about which he is so confident. If so, it does not seem to matter very much whether we assent to his proposition or not. Where is the sense of caring which of two contradictory epithets is chosen, in the absence of intelligible reasons for choosing between them clearly ? Perhaps the assertor reminds us that though A and non-A are always found entangled, still we can often find a preponderance of one or the other, and can name the case accordingly. There are heroes, for instance, though no man is purely heroic. But this defence only changes slightly the form of attack, if the critic is bent on pressing the 1 1 8 DISTINCTION question. When the assertor pretends to have struck a balance fairly, the enquiry after meaning becomes a demand to see the items of the account ; and their very existence as balance-sheet items depends on a clear separation of A from non-A. An account where the debit and credit entries are liable to be mistaken for one another, can hardly claim to perform its function properly. Apart from its insinuations, the controversial ■enquiry after the exact definition of a word puts a direct question to the assertor as to the meaning which he intends that word to bear in some particular assertion. The critic ostensibly asks to be allowed to come to some agreement with the assertor as to what is really meant ; in the absence of such agreement (he complains, in effect) he can neither test the assertion by applying it to particu- lar cases, nor make use of it even if he admits its truth. This, at any rate, is the direct force of the enquiry. It is not primarily a question either of the common acceptation of the word whose defini- tion is asked for, or of the meaning which is most convenient or best on the whole, but of what some one particular assertion, as made by some one person and at some particular time, is intended to mean. We shall see further on ' how it is that the ' Pp. 193-199- THE DEMAND FOR STRICT DEFINITION 119 insinuations of the Socratic enquiry are the same as those of the demand for proof At present the point to notice is the result of ambiguity — its effect upon the meaning of an assertion. Some account of this was given in general terms in the preceding chapter, but now it remains for us, with the help of instances showing the use of this method, to see the matter in a more concrete way. The older and simpler examples — plenty of which are to be found in Plato's Dialogues — are not very good for our purpose. They are apt to seem rather too academical, too little in touch with the puzzles that really trouble us. But it is not very difficult to recognise the old method in some of its modern disguises, and the form of it which seems most apt for our present purpose of illustra- tion occurs so frequently, even outside philosophy, that it must be familiar to everyone. When I make a sweeping assertion which may, as the phrase runs, mean anything or nothing, am I under an obligation to make its meaning clear ? Granting that by an obligation is here meant no more than that the assertion will otherwise fail to carry a meaning to the audience, it seems at first obvious enough that I am so obliged. And yet there are many sweeping assertions, true enough to be useful, which, if treated strictly by this DISTINCTION method, can be made to seem foolish or false. An instance, or an applicable definition, is asked for, and the assertor fails to supply one ; upon which the critic more or less modestly assumes the air of having exposed a fraud. As every reader will know by his own experience, this assumption is sometimes unjust ; either the instances, though many in number, have simply escaped from a memory stored with more valuable deposits, or — and this is a still more vexatious occasion — there is in the nature of the case some reason why no quite satisfactory instance, no perfectly applicable definition, is ever likely to be found. Among the typical occasions of this latter difficulty are those where a ' tendency ' is asserted, as happens largely throughout all science, and perhaps most strikingly in the science of eco- nomics. There was a time, which ended some twenty years ago, when most economists conceived it to be a part of their duty to apply the econom- ist's ' laws ' directly in the solution of political and social problems. They were inclined to assume, for instance, that obstacles to freedom of trade, or freedom of contract, were in all cases simply to be condemned ; and they supposed that the notion of a ' wage fund ' could be used to limit all hopes of obtaining better wages, otherwise than by a decrease THE DEMAND FOR STRICT DEFINITION 121 in the number of labourers. By thus over-simpli- fying the practical problems, and by adopting a tone of somewhat impatient dogmatism in conse- quence, they succeeded in bringing the science into a state of discredit among practical men, from which it has hardly yet recovered, although a great deal of care has lately been taken by economists to reduce the claim which is made by the economical laws. These laws are statements of tendency, we are now very often reminded, not rules of action, nor even uncounteracted uniformities in Nature. In other words, they are abstract laws, and as such they are ' almost never by themselves decisive.' Before arriving at a conclusion in actual cases, we require to know whether, in the given case, the abstract tendencies are anyhow counteracted.' That is, in fact, the essential difference between « an abstract law and a law (i.e. a statement of uni- formity) which is other than abstract : the former ' In this connection it may be noticed that in economics the definitions used are very commonly ' inapplicable ' ones, much to the discomfort of the student who prepares to answer the question whether so and so is wealth or not, or which of some given list of things are capital. Sometimes there is a reference to intention or expectation, as when capital is defined as that part of a man's stock which he expects to afford him a. revenue ; sometimes there is a reference to results without fixing a limit of time, as when pro- ductive labour is defined as labour which increases the sum of human wealth. 122 DISTINCTION may be universal in spite of effectual counteracting causes, the latter ceases to be universal if a single exception can be found ; so that an abstract law may be true in spite of apparent failures in its verification. And the practical difficulty hence arising is that of testing the truth of the law, and of seeing how far it is really applicable in actual cases. Like any universal assertion — like ' all men are fallible,' for example — an abstract universal appears to state the conditions of an inference, and to apply it at all is to interpret it as doing so ; but, unlike the strict universal, it states the conditions of the inference so vaguely that we cannot in practice apply it with perfect confidence — with such confidence, for instance, as in the inference from human nature to fallibility. To say that where we find A we may infer C, when we only mean that in the absence of counteracting circumstances, or ' other things equal,' the inference holds good, is to state the conditions of the inference vaguely. The condition named A is not, in such a case, the only relevant one ; so that the conditions as a whole are only described roughly and incompletely by such a name. To make the statement literally true we should have to introduce further limitations or qualifications — as, for instance, ' A, when B is present (or absent)' — some details which should THE DEMAND FOR STRICT DEFINITION 123 define the requisite conditions rriore closely. And the progress of science always shows itself in some increase in our power of specifying more completely the essential conditions of inference in given cases. A scientific law is revised and improved so far as "we learn to bind up its exceptions along with the general statement. For instance, a red sky at night means a fine day to-morrow — usually, but not always ; if we knew more of the causes at work, more of the reasons for the rule and its exceptions, we should be able to specify more distinctly the trustworthy signs of the coming weather. Take, for example, the assertion that ' popula- tion tends to outrun subsistence,' and remember the treatment it has actually received at the hands of some of its opponents.' It does not concern us here to ask whether in this or that case the critics are right or wrong, but only to notice why they think they are right. None of them, I believe, deny that in a certain abstract and hypothetical sense the law is true ; what they seek to deny is its practical value, its utility in enabling us to foresee concrete events. They take an extended view of the facts of the world, and find that the • E.g., Mr. Henry George in Progress and Poverty; or, for a more sober attack, Prof. Rickards in Population and Capital. 124 DISTINCTION abstract tendency of population to outrun sub- sistence has, and has always had, so many forces actually opposed to it that somehow it has never yet caused serious trouble, except locally and in an accidental way,' Hence, they suggest, it stands on much the same footing as the equally undeni- able abstract tendency of men to live for ever ; the result, in both cases, is ' habitually prevented from occurring,' and for practical purposes we want to know rather what is likely to occur. Another example of the demand for an applic- able definition may be found in Cairnes's objection^ to Jevons's (and Say's) definition of utility. Here Cairnes confesses himself 'wholly unable to conceive how ariything amounting to a real explanation can be extracted from a proposition which,' as he sharply puts it, ' amounts to this and no more — that value depends upon utility, and that utility is whatever affects value.' The name 'utility,' he further explains, is here given to the aggregate of unknown conditions which determine the phenomenon, and then the phenomenon is stated to depend upon what this name stands for. ' Suppose, instead of ' ' The globe,' says Mr. George, ' may be surveyed and history may be reviewed in vain for any instance of a considerable country in which pressure and want can be fairly attributed to the pressure of an increasing population.' Leading Principles of Political Economy, pp. 17-21. THE DEMAND FOR STRICT DEFINITION 125 " utility," we call the unknown conditions x, we might then say that value was determined by x ; and the proposition would be precisely as true, and, so far as I can see, as instructive, as Mr. Jevons's doctrine. In either case the information conveyed would be that value was determined by the condi- tions which determine it, an announcement the importance of which ... I must own myself unable to discern." There are certain people, and the late Professor Cairnes was one of them,^ whose vision is keen in detecting the fault called ' truism.' And in fact such keenness, just or unjust, is extremely useful in con- troversy ; the accusation so brought is one of the most effective forms that the old Socratic method takes in modern times. We frequently meet with examples of it in common life — ^for instance, where some general rule is repeated to us when our enquiry really concerns the exceptions to that general rule. Maxims of advice are especially liable to be thus received, to be classed as ' copy- ' It is perhaps worth noticing that the use of this definition of utility may conceivably be justifiec} on other grounds than as an ' an- nouncement ' of fact at all. If so, Cairnes's objection would fall as flat as the objection sometimes brought agdnst a. theory that as a dogma it will not stand. See p. 187. ' See, for instance, Leading Principles, pp. 29 note, 51, 79, 102, 151. 126 DISTINCTION book wisdom ' ; the adviser seems either himself to lack insight into the real difficulty of the case, or to underrate our powers of seeing it. The stand- ing conflict between the world's opinion and that of the individual may in every case be resolved into a charge of stating truisms, a charge rightly or wrongly brought by the individual against some piece of advice. Worldly, maxims, like proverbs; are conceived to fit the average, rather than the special, case ; the individual may admit the world to be ' wise ' without supposing it to be infallible. Most of the instances of this conflict provide, indeed, such warmly disputed matter that it is hardly safe to use them for illustration, but there are a few instances where the value of the world's opinion can be definitely tested in a short time ; thus, the market price of investments is a case in point. The price of a given stock at a given time is the closest meaisure we can get of the world's opinion as to its present value, and is a far closer measure, by the way, than we ever get of the world's opinion on any subject where quantities are not concerned or where the interest in the question is rather pro- fessed than real. Yet to a given individual the market price may often rightly seem too high or too low. Consider the feelings of such an investor when solemnly told that ' the higher the interest THE DEMAND FOR STRICT DEFINITION 127 the greater is the risk.' The intelligent investor so advised is exactly in the position of the in- dividual to whom a worldly maxim is given in a case which to him seems to depart importantly from the average. The maxim is well known to him, and he firmly believes in its general value, but in the special case it has become irrelevant to his own enquiry. His own enquiry is not ' what is the general rule,' but ' does this case belong to the admitted general rule or to its exceptions ' ; and so to him the mere re-assertion of the rule has no relevance, has absolutely no meaning in regard to the point at issue. The charge of stating a truism occurs, of course, chiefly in connection with very general state- ments, for the more a statement aims at compre- hensiveness, the more likely it is to be either a truism or else untrue. This dilemma is, therefore, a common one in philosophy, where the aim at comprehensiveness is a source of such endless trouble. For instance, the axiom as to the uni- formity of causation is difficult to state in any way which shall avoid the horns of this dilemma. That the same cause is invariably followed by the same effect is perfectly true, if we admit that we are liable to be misled in calling two actual events ' the same ' before we have seen their results, and that 128 DISTINCTION therefore we cannot use the axiom with perfect confidence to predict this or that event in the con- crete ; but false if we mean that events approxi- mately the same (or that seem the same on the surface) will always be followed by approximately the same consequent events. Every mistaken prediction that we make is due to our believing too easily in the applicability of this axiom in its true form, or in its truth when stated in its applicable form.^ And Mill's celebrated 'Inductive Canons'^ exemplify the results of confusion on this point. Every one of these is made true, and yet practically inefficient for prediction, by the presence of an 'if ' If only one circumstance has varied,^ says Mill, ' that one is the cause of whatever further variation occurs.' Of course it is, but how are we in practice to be sure that only one circumstance has varied ? And, to begin with, what are the limits of a ' single circumstance ' ? The real difficulty in induction, the problem how to observe the relevant facts correctly, is just the problem that Mill's canons ' I dp not mean that this axiom is useless altogether ; only that it is useless for prediction in actual cases. There is a negative form of it which may be of considerable use in explanation, after the event, viz. 'Non-uniformity in the consequent points to non-uni- formity in the antecedents,' or 'When we find two consequent events at all different from each other, we may be sure that there was some difference in their antecedents.' ^ See Appendix, p. 261. THE DEMAND FOR STRICT DEFINITION 129 suppose to be previously solved. Regarded as practical guides, therefore, their effect is to direct our attention away from the practical difficulty, to minimise that and make it appear too simple. Their wisdom is undeniable, but it is the kind of wisdom that becomes unmistakable only after the event. These examples, I hope, will be sufficient to show the force of the complaint that an assertion is ambiguous and so loses its meaning entirely. A truth must be capable of concrete application somehow, if it is either to be used at all or tested in the light of apparent exceptions. A truth that we can neither test nor apply in practice lacks meaning as much as if it were in an unknown tongue. It is very convenient no doubt for the assertor, the critic hints, to state his ' abstract law ' so that no apparent failures in its verification can affect it, but what the world really wants is some- thing else than this. Truth, to be worthy of the name, must (the critic says) be verifiable by refer- ence to concrete facts. Assertions whose truth is only saved at the expense of their applicability are, he says, in this dilemma — interpreted one way they are undeniable but meaningless ; interpreted in any other way they cease to be true. No amount of certainty that X leads to Y will help us K I30 DISTINCTION in foretelling the occurrence of Y, if X is a term so ambiguous that we cannot be sure when we actually have a case Of it. And, similarly, no amount of certainty that S is P will give Us more than a merely verbal knowledge of the character of S, or enable us to predict behaviour in given cases, if the line between P and non-P is vaguely drawn. For then our assent to the statement that S is P becomes only a preference for one unmeaning form of words instead of another. If assent and denial are not clearly marked off from each other in their meaning, how can it really matter which of the two forms we adopt .■' When skilfully used, the demand for an applic- able definition is a formidable weapon in con- troversy, but in the following chapters I shall try to show how ft may on occasion be rendered harmless. CHAPTER XI THE ' SPECIAL OCCASION ' Unreal distinctness, as we saw in Chapter V., is only ' unreal ' for those who find it so. The very recognition, besides, of any distinction as rough, implies that we are in a. somewhat two-sided or self-contradictory state of mind as to its roughness ; so far as it seems to be a distinction at all, the dis- tinguished things seem clearly and sharply distinct, while so far as it seems to be rough their distinct- ness seems ' unreal.' More than one explanation might, no doubt, be given of the manner in which we thus come to believe in distinctions and yet not to believe in them — not to take them quite seriously. That is to say, the cause of the incon- sistency is doubtless different in different cases. Sometimes, for instance, we merely forget oUr own admission of the roughness ; that admission was made, perhaps, when we were on our best reasoning behaviour, and the heat of contro- versy or the need for relaxation has since brought 132 DISTINCTION us back to our ordinary state of mind. Sometimes an idle avoidance of trouble will make us put refinements of thought out of sight. But some- times, on the other hand, our artistic sense, or our moral sense, or our practical sense, is offended by casuistry ; we perceive, or think we perceive, that doubts and refinements are in the particular case irrelevant or confusing. It is this that forms the only sound excuse for unreal distinctness. Some- times a rough distinction is sufificient ; that is, in brief, the whole story of justification. But sufficient for what ? For most purposes or for one .■' This notion of sufficient distinctness, or of the irrelevance of the demand for a definition, is in itself somewhat ambiguous, and we have now to distinguish between two very different interpreta- tions of it. Though our chief business in the preceding chapters has been to obtain a general view of the extent of unreal distinctness that exists in our words and notions, and of the manner in which harm may be done by it, yet already several hints have been incidentally given as to the answer which still remains to be found to our problem as a whole. It was partly in order to lead up to this, for THE 'SPECIAL OCCASION' 133 instance, that the nature of the conflict between philosophy and common-sense has been so ex- pressly brought into notice. If that conflict means anything, we have now sufficiently seen it means the mutual adjustment of claim between two oppo- site methods of judging, each of which has much to say for itself — the method that loves a general rule, and the method that loves an exception. These rival ideals, like other rival ideals, are com- monly set in artificially sharp opposition ; we dignify as common-sense, or worldly wisdom, our mere impatience with exceptions, or we dignify as philosophy our puerile fancy that truth is only to be found in paradox. But we have here chosen to conceive the opposition less sharply and easily, and are therefore prevented from setting either method entirely above the other. It is not, indeed, any part of our purpose to attempt to map out the two provinces completely, but only to guard against certain encroachments on one by the other. And if we simply try to allow both methods their fair share in solving the line-drawing difficulty, that will involve our paying some express regard to exceptional cases and special occasions. Though the distinction between common-sense and philo- sophy is a gradual one, the higher grades of either are always ready to bring against the lower grades 134 DISTINCTION exactly the same objection — slavish adherence to rule — that morality brings against law, or religion against a narrow and formal morality* The old conflict between worldly wisdom and childlike directness of vision would not possess so much vitality as it does but for the fact that the child, like the scientific man or philosopher, is often less hampered by convention, or prejudice, or cut-and- dried rule, in observing special cases. But a much more plain and direct hint has also been given several times in these chapters. For instance; near the beginning ' we noticed that am- biguity is- most troublesome where the things or cases that a descriptive name groups together resemble each other closely but yet differ in ' deeper or more occasional ' ways ; or where a word in much of its -everyday use is plain and unmistak- able, and only becomes ambiguous ' on rare occa- sions ' ; and that on any given occasion of the use of a class-name the question is liable to • arise whether the resemblance or the difference is the more important. And again, near the end '^ of Chapter I., the promise was made that we were to see, in the sequel, how casuistry — that is to say, the Socratic demand for a definition — may be on occasion defeated through irrelevance to some ' P. 5. ^ P. 12. THE '■SPECIAL OCCASION' 135 ' special and passing purpose.' And again, towards the end ' of Chapter VI H- it was said that the practical problem is to distinguish with increas- ingly greater correctness the relative importance of differences in regard to ' a constantly shifting and various set of occasions.' This notion of judging the value of every distinction by reference to the special occasion of its use is the point at which we begin to quarrel with unphilosophical common-sense. Even here, indeed, the quarrel need not be a very severe one, since, so far as common-sense really exercises the ' tact ' which it claims to possess, it is adopting the very same method as philosophy, though less con- sciously. The exercise of tact involves occasional and judicious departure from hard-and-fast rule, and our quarrel, therefore, is not with common-sense as such, but only so far as (through hurry, or fear of paradox, or lack of discriminating power) it does in fact allow some general rule to keep our thought in chains. And, since each party to any dispute about ideals accuses the other of being misled by a faulty distinction, no one can well deny that such failures of tact, or of patience, in interpreting dis- tinctions do sometimes occur. Indeed, the relative importance of differences is often not judged by ' P. 96. 136 DISTINCTION means of any reference, either conscious or uncon- scious, to the occasions of the use of the word that hides them, but an easier plan is followed ; common- sense tries to cut the knot by grouping the various occasions as well as it can together. Differences that are important on most occasions, or on most of the more important occasions, are for such common-sense the ' important differences.' Here, as throughout its procedure, exceptional cases are disliked. An example of this is provided wherever the charge of quibbling is brought against anyone in the name of common-sense. For instance : — Lord R. Churchill: ' He says it is well known in war that movements which are offensive in their nature are sometimes defensive in their essence.' Mr. Gladstone : ' Offensive in their form.' Lord R. Churchill: ' What does that come to — that the attack of General Graham was offensive in its form but not in its nature ? Three thousand men, or more, were slaughtered, as a matter of form, by movements which were not offensive in their nature ! ' Here Mr. Gladstone's critic claims to speak in the name of common-sense, and so objects to his somewhat exceptional use of the word ' offensive.' One of the best ways in which the rivalry of these two methods may be understood, is by refer- ring to the familiar difference between common- THE 'SPECIAL OCCASION' 137 sense and science on the question as to what con- stitutes essential resemblance or difference. This question is of great importance in all reflection upon our judgments, and the phrase has passed into everyday use. There is hardly, a person, able to read at all, who does not know its meaning to some extent ; everyone is glad of so convenient a phrase. When we think we see an analogy between two things or events, A and B, we call them ' essentially the same,' and when we dispute someone else's opinion that A and B (say, the Suez and Panama canals) are analogous, we call them ' essentially different.' And, however superior unscientific opinion may be to scientific opinion in the actual practice of recognising essential resemblance or difference in this or that case, yet the general account it gives of its own procedure is unsatisfaptory ; for it shows a tend- ency to speak broadly of degree of resemblance and difference, and to judge whether a given case of resemblance or difference is great or small rather by means of a rapid survey of its salient features than by carefully analysing it into details of greater and less importance in regard to some special purpose. Now, there is no doubt that the vcitories of science have been won by exactly reversing this procedure ; by. cultivating a steady 138 DISTINCTION distrust of the resemblances and differences that are most obvious and striking, in favour of those that, though small in appearance, have an especial im- portance in regard to some question that happens to be in hand. It is rather the scalded dog than the scalded man that fears cold water ; the scalded man long ago set his wits to work to discover exactly what the condition was upon which the scalding depended, and he found that, however great may be the degree of resemblance between hot and cold water, yet the difference, in spite of being less obvious, is essential so far as scalding is concerned. For the purpose of quenching fire, or diluting another liquid, and so on, the difference between hot and cold water is unimportant, or nearly so ; but to call it, therefore, unimportant throughout would be a procedure too rough-and- ready for any common-sense above the level of a dog's. It is here suggested that the question as to the importance or unimportance of roughness in a dis- tinction is fundamentally a case of this kind. Just as science, in estimating the relevance of details, comes into conflict with unscientific opinion, so philosophy comes into conflict with common-sense in estimating the relevance of the objection that a distinction is rough. 'Essential' resemblance or THE 'SPECIAL OCCASION' 139 difference is at bottom ' important ' resemblance or difference, and importance is always a matter of rele- vance to some purpose, and purpose is always liable to vary with the special occasion. The easiest plan of judging importance or relevance. — if we do not care very much for correctness of judgment — is to regard * importance ' as an amount, and to judge the amount on general grounds somehow, or by means of a hasty review of its obvious and striking features. Thus, in judging whether an objection to some rough distinction is important or not, we are always tempted to think of its general import- ance, broadly conceived — instead of its special importance, its relevance to the matter that happens to be in hand. And the notion of (what may be called) ' general relevance,' though not to be entirely discarded, must be superseded by that of special relevance wherever the two conflict. Its value, though real, is secondary ; our knowledge of where irrelevance commonly begins may help us in judging whether and when an actual enquiry is justified ; but, after all, it is the actual enquiry that wants a justification, and this or that actual enquiry cannot be finally condemned on the ground that somewhat similar enquiries (or even the same enquiry) on somewhat similar occasions, are trivial. Thus the difficulty of distinguishing between man I40 DISTINCTION and beast, though it destroys the meaning of the question {e.g^ whether' man invented language or had it from the first, leaves us able to speak unambiguously about existing men as distinct from existing beasts. In short, when comrnon-sense claims that a contrast is ' sufficiently ' sharp, or complains of the ' irrelevance ' of an enquiry into the exact definition of a term, it is apt to forget that both sufficiency and relevance are relative to some purpose ; that there is no such thing as sufficiency or relevance at large. The meaning of the plea, as it is commonly made, is that the borderland between A and non-A is small in amount (orextent), and therefore unimportant ; • the only sound plea is that, whether small or large in amount, the roughness does not at all aifect some purpose immediately in view — that the roughness, though it has relevance in other connections, has absolutely no relevance at all between the assertor ' Did ' man ' invent language, or was it an original possession of the human race ? The question assumes that the two alternatives exclude each other ; but if the line between man and man's non- human ancestors be broken down, how are we to answer either ' yes ' or 'no' to either portion of it ? Why should not the answer accept both the supposed alternatives, and also deny them both ? The first persons who used language were either m^n or men's non-human ancestors, whichever we choose to call them. And in the question as stated the distinction between inan and not-m^n is conceived too sharply for the purpose for which it is used. THE 'SPECIAL OCCASION' 141 and his audience on some special occasion or for some limited and passing purpose. We shall return to this subject presently,' and I shall try to show the truth and bearing of the above remarks in the light of examples. Mean- while, in order to help in obtaining a concise statement of the result we are to reach, it may be useful to refer back to what was said * about the practice of naming ' unrealities ' — naming, as if they were real, things which admittedly have only a potential existence. For this is closely analogous to the practice of using a distinction which we admit to be not quite sharply applicable. Just as, in the one case, we leave out of sight, for a passing and limited purpose, the doubts affecting the actu- ality of the 'thing ' which is named, so in the other case we leave out of sight, for a passing and limited purpose, the real difficulties that we know to starid in the way of drawing the line. In both cases there is an assumption, or a postulate and a concession, that these questions of detail may for a time be considered irrelevant. And it would help us greatly in clearing our ideas about distinc- tions generally if we had some way of describing their nature so as to emphasise the fact of their dependence, not on bare difference, however great ' P. 150. ^ Pp. 61-69. 142 DISTINCTION or natural, but on essential difference, however small or artificial — on difference the importance of which varies for different purposes ; so that any special occasion may on the one hand require that some real difference should be put as far as possible out of sight, or, on the other hand, compel us to notice a difference which our ordinary language leaves unrecognised. CHAPTER XII THE RELATIVITY OF DISTINCTIONS Stated in its shortest and most general form, the truth which we are seeking to establish is that the validity of any distinction is relative to the purpose for which it is used at the time, or that the question whether a given distinction is valid can be decided only by reference to the purpose for which it is used on a special occasion. But these short statements will require to be expanded. By the ' validity ' of a distinction is here meant its resistance to criticism, its right to escape con- demnation on the score of inapplicability, or unreal distinctness. The distinctions we use are ' valid ' so far as our words are free from ambiguity. After what has been admitted already ' about the con- tinuity of Nature, we can make no plausible pre- tence of disputing the charge of unreal distinctness itself ; every distinction is rough if we choose to be strict in demanding applicability ; the charge must, ' P. 71* 144 DISTINCTION therefore, be admitted and yet somehow disarmed, if we are to avoid the deadlock into which the con- tinuity of Nature at first appears to lead. In other words, we must find some way of answering ' true but irrelevant ' to the complaint that the actual line between A and non-A cannot be drawn exactly, and hence it was that the notion of relevance came to be discussed in the preceding chapter. And it follows from what was there said that the notion of ' special relevance ' and that of the ' purpose ' for which a distinction is used are each involved in the other. For instance, the purpose of the distinction between hot and cold water may be on occasion either the question as to the causes of scalding or one of the other questions where the distinction has less importance ; and the purpose for which the distinction between man and beast is used may be either {e.g) the question how far natural impulses should be discouraged, or the ques- tion how far thought is dependent on language. Such purpose is always, in the end, an argument. For distinction is the creation (or recognition) of alternatives, and if these are created (or noticed) for any purpose at all, this can only be that they shall be used as alternatives — that is to say, that we may be able to argue at least from the assertion of the one to the denial of the other. Where two THE RELATIVITY OF DISTINCTIONS i45 or more alternatives are supposed to exist, whether alternative epithets {e.g., good, bad, and indifferent), or alternative courses {e.g., take it, or leave it), or alternative explanations {e.g., on purpose, or by acci- dent), the point or meaning of their separation into distinct alternatives lies in the distinctness of that separation. If that be lost — as happens where the charge of roughness is relevant — the alternatives fail to fulfil the purpose for which they were de- signed. Hence, a phrase that will sometimes be of ser- vice in this connection is the ' argumentative use ' of a distinction. A distinction is used argumen- tatively where the assertor lays stress, or emphasis, on it ; where he wishes to use the terms A and B, or A and non-A, as distinct, and so takes upon himself responsibility for drawing the line between them. If we use this phrase, however, we must remember that it is only so far as the assertor intends to found an argument of his own upon the name A (or non-A) that he takes any responsi- bility for its definition ; and, as we shall see later,' to adopt a distinction ' for the sake of argument ' is a very different matter. Yet in spite of this the phrase is a useful one, as reminding us that some ' P. 187. 146 DISTINCTION argument, whether of the assertor's own or not, is the only purpose that ever explains a distinction — explains, that is, our recognition of this or that difference while other differences are overlooked. When we group men on one side of a line and animals on the other, we overlook for the time the difference between one man and another, or between one animal and another, fixing attention on the dif- ference that for some purpose or purposes, however vaguely conceived, we regard as more important. And it follows, further, that all useful criticism of the names or distinctions used by the assertor resolves itself into the judgment that his argumentative use of them is in some way unjustified. He may make what distinctions, or use what names, he pleases, so long as he does not seek to lead us by nieans of them into conclusions which we decline to admit ; it is when he attempts to do this that we begin to raise awkward questions as to the stress which his distinctions will really bear. But we shall best begin to see the meaning of the short statement which was made at the be- ginning of this chapter by asking what other doc- trines it contradicts or supersedes ; and though- a few hints as to this may be gathered from what has been said about the opposition between the com- mon-sense and the philosophical methods of esti- THE RELATIVITY OF DISTINCTIONS 147 mating relevance, yet there will be some use in raising the question more directly. The task, however, of proving that common- sense really holds this or that false doctrine is far from an easy one. Besides the difficulty of decid- ing who are the authorised spokesmen of common- sense, that ' tact ' which we have already mentioned so often enables those who possess it to elude the logical critic in a very slippery manner. It is always easy to claim that your general statement (if you take the trouble to make one at all) is only broad or rough, and that you yourself know all its faults, and never dream of taking it in its bald and literal interpretation. It is meant to be used, you say, just as sensible people use a proverb ; it pre- tends to no more strict generality than statements like ' where there's a will there's a way,' and similar incomplete truths. So that even if you assert, for instance, that the validity of a distinction is best judged by means of a rapid comprehensive glance or a happy instinct that refuges to pay any peddling attention to subtleties, you may in the end be only describing with some artistic license a process which you really perform in a sufficiently capable manner. And this is perhaps an excellent defence on the part of the common-sense individual against criticism on the part of a philosopher. That is to 148 DISTINCTION say, on the face of it, and apart from chance oppor- tunities of getting behind the scenes, it leaves the philosopher with no firm ground for a personal accusation. But, on the other hand, it is possible that the personal accusation is not the true centre of interest ; the philosopher, in finding fault with some general statement, may not have had in view the case of the gifted individual who interprets it safely, but rather that of the many less gifted indi- viduals whose outlook, already narrow, is still further limited by the prison walls of the doctrine in ques- tion. He will then admit the analogy between such a statement and proverbs, but will point to the fact that proverbs also exercise an extremely misleading influence — an influence that would be much worse than it is but for the inconsistencies and contra- dictions that opposite proverbs encourage, so that what the ignorant learn from one they can unlearn from another. With this explanation, however, it may be said that the background of the doctrine that distinction is relative to special purpose— the opposite doctrine which is contradicted by it, and by contrast with which its meaning stands out — is that the validity of distinctions is at any rate something that does not come and go with the context, but inheres in the distinction somewhat as a good constitution THE RELATIVITY OF DISTINCTIONS 149 inheres in a healthy man ; that distinctions are to be classed as more or less rough on the whole, or even as rough and non-rough, without reference to the special occasion on which they are used ; in short, that a rough distinction is always rough, and a sharp one always sharp. This common-sense doc- trine, like a proverb, is an incomplete truth which in practice misleads people easily. Though we can seldom say beforehand for certain that it will mislead this or that person, it is generally possible to discover afterwards that it has done so ; and perhaps the best way to contrast the influence of the two opposed views respectively will be to examine a few examples where a difference of opinion has arisen as to the validity of some dis- tinction. These, therefore, will be found in the next chapter. CHAPTER XIII SOME ILLUSTRATIONS The question whether a distinction is valid takes in practice so many different forms that it is dif- ficult to select any one of them as fairly represen- tative. In fact, if it be true that all criticism of judgment is criticism of distinction,' it follows that wherever opinions conflict the question is raised, however obscurely, whether some distinc- tion is valid. We are, therefore, free to take any examples of divided opinion and use them for illustration. And what we want especially to notice are the consequences of failing to see that the validity of distinctions comes and goes with the occasion. Such consequences are of two opposite kinds, the commoner ^ kind being exces- ' See page 193. ' It is generally easier to leave a distinction uncriticised than to see it as fluid and abstract, and that is why I suspect the former to be the commoner fault. But the question of relative frequency is only of accidental and local importance. If our tact in the use of distinctions fails in either direction, the result is equally error. SOME ILLUSTRATIONS 151 sive belief in the validity of some distinction which has, for the moment or more, become invalid ; and the rarer kind being refusal to take (even for a moment) some distinction as valid whose faults have once been seen. Certainly, one result of failing to see the rela- tivity of distinction in general is that distinctions whose faultiness only creates a difficulty on rare occasions tend in practice to escape criticism alto- gether. It is very natural that such distinctions should be accepted uncritically, classed as valid once for all, and no further thought be given to the matter ; so that when the rare occasion comes, we are unprepared to meet it. We acquire a habit of trusting the distinction, and in time the habit becomes too strong to allow our critical powers fair play. From the commonest thought upwards to the most abstruse and intricate, this tendency may be seen at work ; and the only difficulty is to select, from among the host of examples daily met with, those that will best serve for illustration. At the lower end of the scale we may place the neglect to criticise such a distinction as that between metaphorical and unmetaphorical language. Within broad limits we all know very well what metaphor is ; it is the salt of expression, and you can have too much of it or too little. In a large percentage 1 52 DISTINCTION of cases the difference is clear enough. But even a very sHght acquaintance with the nature of language will suffice to show that metaphorical words become unmetaphorical by degrees, so that at all times some words are in the transition stage. As a matter of fact, nearly ' all words have plainly been metaphorical when first applied to their present uses, and even those that are most straight- forward to-day have only gradually lost their meta- phorical character. The word character'^ is itself an example. Each of them has for a time remained on the borderland between metaphor and straight- forward meaning ; hundreds (like borderland or straightforward) have a tinge of metaphorical meaning now ; so that it is only a careless view which can content itself with finding no difficulty in applying the distinction. And how, for instance, will the problem as to avoiding assumption under cover of metaphor be dealt with by anyone who sees no difficulty in it .■' We are all at times liable to take metaphorical phrases for direct ones ; much more, therefore, those who do not suspect their own liability to be misled. And the same with the ' We may say, rather, all, except perhaps the names of some of the longest-known and simplest material things, and a few acci- dentally invented names like gas, though even there some already- existing word (gheest, for instance) has often influenced the inventor. ' From x^foKT^p, ' an engraved mark. ' SOME ILLUSTRATIONS 153 merely literary problem as to avoiding a mixture of metaphors. The difficulty is not to be solved by refusing to see it, but needs rather wariness and some knowledge of the ways in which a word may keep or lose, for a time, its metaphorical character. For, apart from the change which comes through lapse of time, context also has some influence in bringing forward or concealing the remains of metaphorical meaning. Hard-and-fast distinctions are the bane of psychology, as every student of that difficult subject knows ; but in popular talk about the mind and its operations or its ' faculties ' very little trace of such knowledge appears. Not only are distinctions, like those between conscious and unconscious mental life, or voluntary and involuntary .action, used with ex- cessive confidence, but other more pretentious dis- tinctions, like that between sensation and perception, cognition and recognition, simple and complex, sub- jective and objective, occasionally find their way also into common talk. In these and very many similar cases language, with its false simplicity, is ever, ready to entrap us into error, and often succeeds in doing so. If we fail to recognise the roughness of such distinctions our conception of the mind and its work will be a mechanical one, as if the various parts of the mind were put together 1 54 DISTINCTION like the parts of a machine. There may, no doubt, be purposes for which such a conception has a temporary value — there is, probably, no fault of language that has not some such justification — but it is when we proceed to build upon these distinc- tions, as if there could be no doubt about their validity, that the effects of the error are seen. Mr. Ruskin, for instance, may perhaps be justified in writing : — 'I wholly deny that the impressions of beauty are in any way sensual ; they are neither sensual nor in- tellectual, but moral.' His intention may merely have been to assert, in an emphatic manner, that common opinion has hitherto rather overlooked the moral element in our vision of beauty ; but if the young disciple should take the sentence quite literally, he would get a much too limited idea of what the impression of beauty really involves. Akin to these lapses in psychology, and at a rather higher level, a frequent common-sense error is that of laying too much stress on the distinction between theory and fact ; that is to say, between deniable and undeniable assertions. Theories, it is supposed, are built upon facts as a house upon its foundations ; facts are said to upset theories as if they were opposed to them and SOME ILLUSTRATIONS 155 wholly independent. As everyone admits, the distinction has often great convenience. Any logic — any theory of criticism — must use it largely, since the main business of such criticism is to distinguish between what is deniable and what is undeniable in any given conclusion. Logic is compelled, therefore (though in a more transient manner the less elementary the logic), to assume that fact and theory stand in opposition to each other. The difficulty consists in duly recognising their opposition without exaggerating its extent, and there is small cause for wonder that we should often fail to solve so delicate a problem. A common form of the error may sometimes be seen in the process of proving ' facts ' before a court of law. But the same liability to confuse, in practice, the deniable with the undeniable, under cover of the ideal distinction between them, occurs in a subtler way wherever we try to separate our theories from the facts they rest upon ; since ' fact ' is always partly theory. All fact, that is to say, is fact as theorised — as seen through spectacles of theory ; so that when we assume any fact to be pure, we do so only by leave, for the sake of turning atten- tion to the use that is made of it rather than to its own possible shortcomings as a fact. The fluid nature of the distinction between 156 DISTINCTION fact and theory foixes itself upon our notice most effectively in the failure of all our attempts to understand in the light of that distinction the earliest beginnings of knowledge. We do not become conscious of any theorising until we have already done a good deal of it unconsciously. We cannot find, or even imagine, the perfectly un- theoretic mind receiving its earliest fact. At the furthest point to which we can get backwards in imagination, if we try as hard as we can to put ourselves in the position of people wholly without experience who first begin to reason about what they observe, we. never succeed in reaching the tabula rasa. The simplest mind we can imagine as mind at all must have at least enough theory, or beginnings of theory, to pave the way for its observation of fact.' How much more, therefore, must the complex and elaborate ' facts ' observed by any grown up person to-day be seen in the light of theory. Our minds are steeped in language, and the use of language with a meaning ^ requires ' This need not mean that observation proceeds entirely on lines laid down by theory. To some extent it may rebel against direction, and thus become liable to an opposite bias. At any rate, conflict with expectation is one of the means by which atten- tion is roused to find fact noticeable ; and expectation is caused by theory. ^ I.e., connotation. See p. too, note. SOME ILLUSTRATIONS 157 a biassed mind. Our least wordy observation of fact only differs gradually from description of it ; though we may think ' without words,' yet just as verbal description implies expressible theory, so this more silent description implies at least theory in a less finished form. In men of science and philosophers, again, the same tendency to exaggerate distinctions may sometimes be seen. For instance, Darwin tells us that, shortly before 1859, 'the great majority of naturalists ' believed that species were immutable productions, and had been separately created ' ; which means at any rate that the distinction between species and variety was then supposed to be a firm one — supposed so even by the great majority of naturalists, and therefore presumably still more by common-sense. And through the great change of belief which has since taken place we can easily get a glimpse of the contrast in result between the earlier and the later view. An ordinary person, at the date when the ' Origin of Species ' was written, might have confessed that he did not know exactly what a species was, as distinct from a variety, but he would have had no doubt that that was merely a piece of ignorance on his part, like • Wallace adds : ' and almost without exception the whole literary and scientific world. ' 158 DISTINCTION his ignorance of the technical difference between, say, a ' crime ' and a ' misdemeanour,' or between the various sorts of type that printers distinguish, or of bricks that are sold in the building trade. In a vague way he might suppose that species were 'created,' but this is only a name for the total absence of theory ' as to origin, and if he ever enquired into their origin seriously, the very sharp- ness of the distinction itself would prevent him from asking just the questions that we ask with greatest interest now. Aware as we are now of the possibility that species are only varieties with the difference somewhat increased and hardened, our attention directs itself to the ways in which this growth might conceivably have come about in special cases : here and there we think we see some links of the chain of causation ; here and there we are puzzled. But at any rate the line of our enquiry is guided now throughout in this direction, instead of being kept away from it by what was supposed to be an insurmountable barrier. The recognition ' I mean, of course, theory attempting to give detailed explana- tion. If we say that species were ' created ' we mean, amongst other things, that we cannot trace any actual steps of the process by which they came to be ; and some anti-Darwinians ^e.g. , the Rev. J. Gerard, see below, p. 187) expressly claim to be on the side of those who ' do not yet know. ' SOME ILLUSTRATIONS 159 of the distinction as artificial opens up a whole new and fertile field of enquiry into facts. Another example may be taken from the Socialist controversy. Socialism ' is supported and attacked on two grounds chiefly — its justice and its expediency. And the Socialist notion of justice, its opponents say, is to a great extent influenced by mere class-hatred, which is not a lofty, and certainly not an impartial, spirit. The Socialist notion of justice is not (the opponents say) com- prehensive enough in its scope, but tends to exalt the hired labourer, and especially the unskilled hired labourer, into a position above that which he really deserves at the hands of the community. For it is not true, in the sense in which Socialists use the phrase, that wealth is ' wholly created by labour ' ; labour misdirected creates no wealth, and the unorganised or badly organised labour of any number of industrious men will produce far less than the same amount of labour when well directed and organised. Moreover, even the best directing ' By this vague term I mean the conscious political aim either at much greater equality of conditions than at present, or at greatly extended regulation and restriction by society at large of the ' rights ' of the individual. These different vpays of describing it are not, however, so different as perhaps they seem at .first. Socialism in all its forms is an attack upon private property. 1 60 DIS TINCTION power, coupled with the greatest industry, may be crippled by want of capital.' This objection, rightly or wrongly, implies that the word labour, as used by Socialists, is ambigu- ous ; that the Socialists fail to see a certain obvious difficulty in defining it ; that they take the distinc- tion between labour and the other factors of pro- duction as a sharp one, and use it argumentatively to prove that hired labour alone deserves the glory and reward. Here we have, then, an instance where the objection is raised that a distinction, useful enough for certain purposes, is used for a purpose where its value holds no longer. The distinction in question is that between labour on the one hand and capital and management on the other ; as productive of value, the economists say, capital and management ' essentially resemble ' labour ; for the special purpose in question they are forms of labour, instead of being opposed to it. The distinction, they say in effect, has thus been put to a strain it will not bear. These examples may be enough ^ to show the manner in which excessive belief in the validity of a distinction disturbs our judgment. Of the oppo- ' I.e., of power to await deferred results. ^ A few instances from philosophy are given in the Appendix, p. 264. SOME ILLUSTRATIONS i6i site error ' we are not ready just yet to notice the full extent ; but one example may serve to show the manner in which it arises now and then. In a review by the late Rev. Aubrey Moore of a book ^ by Mr. Romanes, the following passage occurs : — Mr. Romanes wishes to prove that human and animal psychology differ not in kind but in degree. Here everyone is against Mr, Romanes, including himself, unless he is prepared to say that evolution has abolished species, instead of showing how species came to be. If a cat and a dog are different in kind, so are a man and a monkey, whatever view we may take of the genetic rela- tions of the pairs. But this is not what Mr. Romanes means by different in kind. In a footnote to page 3 he says that difference of kind means difference of origin, and accuses Professor Sayce of ' confusion ' for saying that ' differences of degree become in time differences of kind.' We seem to remember a greater than Professor Sayce teaching us that the categories of quantity and quaUty disappear in ' measure.' And if this sounds to Mr. Romanes a trifle metaphysical, we might remind him that whenever science has shown that differences of kind, considered genetically, are differences of degree, no one dreams of supposing that they are any the less ' Viz. , the case where, through finding a distinction rough, we seek to prevent an opponent from using it, even on occasions where we ourselves are assuming its value. '' Mental Evolution in Man. The review is reprinted in Mr. Aubrey Moore's Essays, Scientific and Philosophical, and the passage quoted is on p. 45. M 1 62 DISTINCTION differences of kind. The question of origin has nothing to do with it. The argument here is subtle, but nevertheless instructive. The difficulty is a familiar one in philosophy, especially in regard to ethical questions such as the nature of the moral sentiments. And what Mr. Moore would have said had he been more careful would probably have been to the effect that where we are speaking only of the fully-developed forms the reflection that A had formerly been indis- tinguishable from B becomes for the moment irre- levant — though even then it is an exaggeration to say that no one dreams of supposing it relevant. Would that common-sense had so much discrimi- nation ! As it stands, however, the meaning of Mr. Moore's argument appears to be to deny the relevance of the objection, on any occasion, that a distinction is gradual. If A and B are found to be gradually distinct, he says in effect, no real difficulty is thereby ever raised as to their distinctness. Man and beast differ in kind, although the difference is only gradual, and ' no one dreams of supposing ' otherwise. If this means anything, it means that A and B can be distinct in kind and yet differ only in degree. It may seem at first sight that the strict truth of this doctrine (assuming that we have found its SOME ILLUSTRATIONS 163 meaning correctly) must on our own principles be admitted. If Nature is continuous throughout, then certainly all differences are gradual, even those that are most specific, and, therefore, the fact that A and B are only gradually different does not in perfect strictness prevent their being also different in kind. How, then, can we, of all people, object to Mr. Moore's conclusion .' We can ask what his meaning is. We can claim with justice that his doctrine has no meaning so long as the name ' difference in kind ' is used at all. For in order to use that name, in order to put any meaning into it, we must have some alternative contrasted with it, and that alternative is ' difference in degree.' There is no point in calling any differ- ence specific except between parties who agree to recognise some line, however artificially drawn, between differences which are specific and those which are not so. Besides, distinction always pretends to set different kinds apart, and so the admission that a distinction is really gradual means that this pretence is not justified in fact, however justifiable it may be on grounds of convenience. In admitting the truth that underlies Mr. Moore's doctrine, therefore, we are far from admitting his incidental assertion about what people in general dream of supposing. On the contrary, everyone. 164 DISTINCTION Mr. Moore himself included, not only dreams of supposing, but inevitably does suppose — whenever he uses the name ' difference in kind ' at all — that its meaning is somehow to be contrasted with that of ' difference in degree ' — not confused with it ; the distinction which is drawn between these two notions, though it is (like all other distinctions) artificial, is made firm for the moment by all who use either name for any purpose. The wider and closer our survey of actual cases of faulty distinction becomes, the clearer to us will be the truth that distinction rests on purpose, and can always be justified (for the moment) — however ' faulty ' it be — by reference to that purpose. No distinction is ever seriously made unless a difference is recognised as having importance ' — importance enough, at least, to be worthy of notice at times. And the recognition that the importance of differ- ences varies with the occasion lies really not far away from our commonest habits of thought. Distinctions among men, for example, have their appropriate times and seasons ; before an English law-court, though not at a State-reception, all men are in theory equal ; we are fully accustomed to regard a husband and wife as for some purposes ' Distinctions based upon unimportant differences appear to be meant when we speak of ' distinctions without difference. ' SOME ILLUSTRATIONS r65 essentially two distinct people, for other purposes essentially one ; and the distinction between a sovereign and a sixpence, important enough in regard to their purchasing power, becomes un- important for the purpose of 'tossing up.' So, again, the distinction between life and death, which is constantly drawn in actual cases with practically clear decision, loses its point where, as in certain lower forms of life, the individual and its offspring are not quite sharply distinct. If an organism is propagated by simple division, which half is the parent ? and when does the parent die ? However vaguely or however clearly drawn any line may be, there are always some occasions for which it is sufficient, and others for which it is not. So vague a line, for instance, as that between eccentricity and madness is sometimes useful ; so firm a line as that between success anA failure is sometimes found to evade a completely final test. To call a distinction faulty never means more than that it is faulty at times ; to call it faultless only means — if Nature be continuous — that for some purpose in view its roughness is irrelevant. Language as a whole may, with hardly a stretch of fancy, be said to be always urging us to lay too much stress upon dis- tinctions — by which is only meant, however, that i66 DISTINCTION we are prone to trust them on occasions when they do not deserve to be trusted. In addition to the hints already given, as to the practical application of these views, it will be useful further to trace their influence more generally. The chief field of their application is controversy, or discussion in the widest sense ; that is to say, discussion whether between opponents one of whom asserts while the other criticises, or between the as- sertive and the critical spirit within ourselves. And we are now to attempt to trace their consequences in regard to the question how an assertor can escape from the charge of trusting too much to a distinction. I do not wish to disregard the fact that the question ' where do you draw the line ' can be made exceedingly awkward for the assertor, but I wish to point out a certain escape that is open to him if he will be content to pay the un- avoidable price. It is in the assertor's natural, but mistaken, dislike to paying this price that much of the strength of the Socratic method lies. As we have seen, the only true escape consists in showing the irrelevance of the demand for de- finition where it is really irrelevant — that is to say. in showing that the admitted roughness of the given distinction does not at all affect some pur- SOME ILLUSTRATIONS 167 pose which the assertor has in view. But of course there are many conceivable phrases which may be used for claiming this, and the one which is here to be proposed will be found, I hope, to be con- venient on occasion. It consists in describing the name whose definition is asked for by an epithet which contains in its meaning a concise claim that the demand is irrelevant. What this epithet is we shall see in the next chapter but one,' and we shall also see what concessions are involved in the claim. ' See especially p. 180. CHAPTER XIV ' PROPER ' AND ' GENERAL ' NAMES The student of elementary logic who has learnt something about definition, and about the ' conno- tation ' and ' denotation ' of names, knows why it would be absurd to ask for the definition of some 'proper' name that occurs in an assertion. He has learnt that if proper names can be said to have a meaning at all, it is a meaning of a very different kind from that which ' general ' names possess ; that the former can, while the latter cannot, be applied to this or that individual case without regard to its nature ; and that consequently the question whether or no they are correctly applied in a given case is to be answered by reference to wholly different considerations. As all agree, the essence of the distinction between proper and general names lies in the fact that the right to a given proper name does not depend upon the pos- session of any qualities that happen to be meant by the name. /*«^er is christened Peter, and rightfully 'PROPER' AND 'GENERAL' NAMES 169 keeps the name although he be soft or unstable ; Bright is born Bright, but dull ; and a name like Sevenoaks or Oxford may survive for hundreds of years the fact which it once described. Proper names, as such, are only accidentally descriptive ; and though it is true that a descriptive meaning can often or nearly always be traced in them, their purpose and value is independent of any such meaning. General names, on the other hand, are always recognised as belonging to this or that in- dividual case conditionally ; a geometrical figure is not a square unless its four sides are straight lines and equal ; an illness is only scarlet fever if and so long as such and such symptoms are present. Always the application of a general name to this or that case introduces debatable matter. The general name is thus essentially de- scriptive ; its applicability is dependent on the existence of certain facts, and properly ceases as soon as the name fails to describe the given case correctly ; when the child grows up, we say that he ceases to be a child. But there is one important fact that, judging from the text-books, the student of elementary logic has small chance of learning, until, at least, the earlier stages of his study are left far behind, and that is, that the distinction between proper I70 DISTINCTION and general names is rather ideal than applicable. If we are asked to say of a given name whether it is proper or general, we cannot always give a decisive answer by mere inspection of the word apart from its context, but we often require to be told how the word is supposed to be used. The distinction, in short, between proper and general names lies not in the words as such — not univer- sally in the form of the words — but in the functions they are made to perform ; and the same word may at any given date be widely and commonly used to perform both functions. When I hear the name Turkey, for instance, or Town, or Bradshaw, nothing except the context can tell me whether the proper name or the class-name is meant. One explanation of this is to be found by noticing the manner in which both proper and general names are usually ' invented.' Since pre- historic times, most names, of either kind, are formed by a process of adaptation rather than by anything that more strictly deserves to be called invention at all. Normally,^ all proper names ' There are, of course, many cases where the general meaning, if it ever existed, is lost ; and a few cases {e.g. the names which children give themselves) where we have reason to suppose that there never was a general meaning. But the normal process of giving a proper name has certainly been that by which schoolboys often invent nicknames, the selection of some marked characteristic, such as ' the measurer ' or ' the shiner ' for the moon. 'PROPER' AND 'GENERAL' NAMES 171 have once been general, and many general names are derived from proper names of an earlier date. If in some cases the meaning of a proper name is difficult or impossible to trace, yet in very many, even of the older ones {e.g. Peter, and Bright, and Oxford ; or, again, Adam and Eve), there is no such difficulty. We may, however, best see the old process now at work in the case of nicknames, and, in general, wherever our choice of a name for a thing is most unshackled by custom. Even in fanciful names like those we give to our dogs, our ships, or our houses {e.g. Lion, the Sea-gull, Ivy Cottage), there is often an attempt to be somewhat descriptive ; in such cases we often select a proper name from among the class of existing general names. As to the opposite movement, where proper names have become general, all language is full of examples. When, for instance, we say that So-and- so is a Croesus or a Solomon, we have taken one step towards the making of a general name from a proper one. Croesus or Solomon, so used, would be descriptive, and would only be slightly less familiar in that use than, for example, the adjec- tives Pharisaical or laconic, the verbs to meander, to lynch, to tantalise, or the nouns rodomontade, martinet, philippic, dunce, &c. Sometimes a class, \n DISTINCTION or kind, of things is called from its maker or finder, e.g. mackintosh, hansom, negus, dahlia ; some- times, like jersey, calico, or canary, from the place whence it was introduced ; sometimes, like quixot- ism., academy, boycotting, from a resemblance to well-known examples in fiction or history ; and in all such cases it is only by accident that we remember for any length of time the origin of the name. The adjective maudlin, for instance, has perhaps lost touch with its origin more completely than the adjective stoical, the substantive solecism more completely than simony, the verb to canter more completely than the verb to jerrymander. The history of the Jill who is supposed to have given her name to the class oi jilts has long ago disappeared ; to the groom, phaeton is only the name of a kind of carriage ; and atlas, to many a bookseller, only means a volume containing maps. But although these examples show that the same word — the same combination of sounds or letters— may often be put to either of the two uses indifferently, yet we shall see the reason more forcibly if we take a wider view of the actual process of using names. We shall find that the same fluidity of usage which destroys the sharp- ness of the distinction between proper and general names, when taken apart from their context, invades 'PROPER' AND 'GENERAL' NAMES 173 also other distinctions that are drawn between kinds of words in grammar and in elementary logic. Look, for instance, at the verb and the sub- stantive. Without going back to any speculations as to whether the verbal or the substantival was the earlier form of the general notion, it is plain, at any rate, that in English there is plenty of give and take between them. Existing verbs, like to give and to take, to grave, to rifle, to lift, give us new substantives ; ' existing substantives, like plunder, or smoke, or wire,^ give us new verbs ; and in words like guarantee, or endeavour, or adventure, we find the verb becoming a substan- tive and the subtantive so formed becoming again a verb.* The same freedom of interchange is " To say nothing of the numerous cases where an English sub- stantive is formed from some part of a foreign verb ; e.g. exit, affi- davit, fiat, plaudit, innuendo, dividend, proviso, restaurant, sou- venir, rendezvous, ozone, phenomenon, &c. 2 In the case of smoke and wire slang has already taken the verbs so formed and has made new substantives of them. We talk of ' sending a wire,' and some people call a cigar ' a smoke.' ' In some cases the to-and-fro movement is really longer. Thus, if we accept Mr. Skeat's account, the word guarantee has the fol- lowing history. From the old High German verb warjan, ' to protect,' came the old French substantive warant or guatant ; from that again the old French verb garantir, the past participle of which became the noun guarantie (or garrantie), and so, in time, the English substantive guarantee ; whence the English verb to gua- rantee was derived. Here, then, we have verb-substantive-verb-sub- 174 DISTINCTION found also between the substantive and the adjec- tive ; a considerable number of adjectives are formed from substantives, not only by means of adjectival endings, like noise, noisy ; ornament, ornamental, and so on, but sometimes ' without any change of sound or spelling, as when monster is used for monstrous, or two substantives are put together, one of which is to all intents and purposes adjectival, as in rose colour, cava/ier treat- ment, drawing-room music, court cards, crown jewels, _field sports, j'udilee coins, &c. And on the other hand a great many substantives are formed from adjectives. Our present English custom does not, indeed, allow us to take any adjective we please and use it by itself as a subject term, but the strictest grammarian has no objection to the ac- complished gradual change of an adjective into a substantive, even without any change in its sound or spelling. For instance, custom at present forbids us to speak of actives as a class, but has no objection to operatives or conservatives ; we may speak of radicals, or of liberals, or of morals, but stantive-verb, without going further back than about the eighth century. • These cases are comparatively rare, though greater freedom in using abbreviations would perhaps introduce more of them. As a rule, in established English the adjectival form, when later than the substantival, is distinguished by the termination. 'PROPER' AND 'GENERAL' NAMES 175 not of financials, or legals, or temporals ; we may speak oi panic, or of logic, or of music, but not of civic ; and a complete list of words, now reckoned as substantives, which /Were properly adjectival at some former time, would be of surprising length.' When we look at the words alone, then — at the words apart from the way they are used on the special occasion — the distinction between the sub- stantive and the adjective or verb is not entirely firm and binding anymore than that between proper and general names. Just as a proper name may become general, or a general name proper, by our using it for one or the other purpose, so we may make a given word either substantival, adjectival, or verbal. In both cases it is the way in which the words are actually used that determines their character. The same is true, as we have partly seen already,^ of the distinction between ' abstract names' and others. No satisfactory account can be given of this distinction by reference merely to the out- ward form of the word. We are sometimes told. ' A sufficient number may be found even if we look only at words of more recent origin, like aneroid, burlesque, cordial, scurvy, stout, &c., and neglect all those — like parasite, missile, uniform, ague, fort, dusk, pauper, sovereign, Serjeant, sloven, &c.— where the change from adjective to substantive was effected in some earlier language than ours. ■' P. 64. 176 DISTINCTION for instance, that abstract nouns are those which are formed from verbs and adjectives by a slight alteration in the form of the word, such as the addition of some termination like ness, or th, or ty, or ce, or hood, or by using the definite article with the infinitive (as in Greek and German), or by otherwise treating the infinitive as a noun. And it is true that a complete list of the nouns by which we name attributes would probably show a majority that are distinguished in some such obvious manner. But there are many other cases — e.g. good and evil, change and decay — where the rule would be insuf- ficient. What is true is, that the formation of abstract names is always a case of using substan- tivally a word which has hitherto had other than substantival uses ; and the fact that we often show our intention by slightly altering the form of the word, is of just the same weight and relevance as the fact, already noticed, that in forming an adjec- tive from a substantive we usually alter its form. It no more touches the heart of the matter than our custom of advertising our births, deaths, and marriages affects the events themselves. What we find is, that just as a given general name may be put to service as a proper name, or a proper name as general, just as a substantive may be formed from an adjective or a verb, and a verb or an adjec- 'PROPER' AND 'GENERAL' NAMES 177 tive from a substantive, so any name whatever may be used to express an abstraction, and after being so used may again be pressed into service to name things that can be seen or felt or weighed. As a matter of fact, not only are all our names for im- material things derived from names which once had a material meaning — the name spirit, for instance, was once, in Latin, the name for breath — but the contrary process is often to be found in operation. We should never have had the word dungeon unless the Romans had had the abstract name dominio, nor dynamite unless the Greeks had had a certain name for the abstract c\UdMty power. These facts, it is hoped, though they lie very near the surface of langua.ge, will be sufficient to remind us that the character of names is not some- thing once for all impressed upon them, but is essentially dependent upon the uses to which the names are at any time put. Words are not, like members of Oriental castes, hopelessly specialised to this or that service, but are able within wide limits to change their trade on occasion. The in- dividual speaker is constantly under greater or less temptation to effect these changes in order to serve a passing purpose ; apd where a large enough number of people find the new usage convenient and not too startling or too little justified, it gradu- N 178 DISTINCTION ally wins its way into general favour. So it comes about that at any given time new uses of words are to be found at all stages of the progress from eccen- tricity, affectation, or slang, to a well-earned place in the dictionary. To admit the fact of this pro- gress, even as only occasional, is to admit, amongst other things, that the received classification of names as ' proper,' ' general,' ' abstract,' &c., is only roughly sound. Ideally, the distinctions are perhaps perfect ; that is to say, they are intended to mark really important differences. But practically there is in each case a doubtful borderland, where great and small experiments in new usage are ever press- ing forward for acceptance, but are not yet fully established by custom. And at the present stage of our discussion there can be nothing strange in this ; it is only one more example of the general truth that our distinctions of all sorts are aimed at clearer differences than the facts of the world will allow us exactly to express by means of them. Taken along with the other examples already given, it may help us to see how often the root or purpose of a distinction is better than its success in dealing with the medley and shifting mass of facts that actually come before us. CHAPTER XV ' REFERENCE-NAMES ' In order to know for certain what kind any given name belongs to, its context is all-important. That is the lesson taught by the facts we have noticed in the preceding chapter. Not the way in which words are formed or spelt or pronounced, but the way in which they are used, is what finally deter- mines their nature ; for words are instruments of assertion, and their own nature varies in dependence on what they are meant to do. For instance, proper names are meant to refer to a subject, not neces- sarily to describe it ; and that is why the demand for their definition is irrelevant. Common-sense, as well as philosophy, agrees that the distinction between proper and general names depends upon what the names are ' meant to do .' The difference is that common-sense, as reflected in grammar, tries to apply the distinction by considering what any given name is usually meant to do ; while philo- sophy tries to apply it with reference to special i8o DISTINCTION occasions. And so we are now to ask what are the cases in which a ' general ' name is really ' proper .' When attempts at this kind of innovation are made, the innovator has to choose between two evils — between using familiar words in an altered sense and inventing new ones. If he takes the former alternative, he runs a risk of producing con- fusion, for old habits of meaning are hard to change ; if he takes the latter alternative — especially in a subject like logic, where every writer feels inclined to take it and many yield to the temptation — he runs a risk of stirring the reader's resentment. I must choose the latter, however, as the least of the two evils, and trust to be able to disarm any just resentment, partly by pleading some reasons for the innovation, and partly by freely confessing the pro- cess to be in itself an evil. We shall, therefore, try, as far as we can, to keep the ordinary rough sense for the words 'proper ' and ' general ' as applied to names, laying no stress upon that distinction ; the distinction upon which we are to lay all the stress shall be that between reference-names and descrip- tive names : by the former being meant all names, proper or general, when and while used for refer- ence ; and by the latter, all names, proper or general, when and while used to describe a nameable thing. In adopting this course we are ourselves afford- 'REFERENCE-NAMES ' 1 8 1 ing an example of the reference-use of names. The words ' proper ' and ' general ' become mere ' reference-names ' so far as we disclaim, in using them, the intention to lay stress on the distinction. Not we, but he who believes in their strict applica- bility is the person from whom a definition of them may be demanded ; for, in answer to such a demand, we have only to say that we do not know and do not care : we do not care what they connote, but only what they denote ; we are using them just as we might use the names ' Peter ' and ' Paul ' — using them to refer to actual cases agreed upon. The moment the agreement ceases we withdraw the name and ask our audience to substitute any other name they please, subject to the condition that an agreement shall be reached as to what that name denotes. When a word is used in this non-committal manner the assertor neither maintains nor disputes its correctness, but treats it just as we treat a proper name. One name, he says in effect, will do as well as another to refer to the case about which we are speaking ; you may call these cases So-and- so if you choose ; I can certainly have no objec- tion to that name so long as you leave it colour- less — so long as you do not use it to prejudice the question raised. For instance, in discussing the' i82 DISTINCTION relations between religion and science, it is quite possible for the two opposite parties to speak of scientific men or religious men in the ordinary sense of these terms, even though the one party may wish to assert in the end that the religious men have the wider and truer knowledge of reality, or the other party to assert that the scientific men have the deeper and firmer religion. The open use of the epithet so-called is only a clumsy expedient for preventing either party from using the name to beg the question. Between people of fair intelli- gence the qualification may generally be under- stood. It is in this way that we may speak of the electric current, though we believe it is not really a current at all ; or of Aj/arro^^^^^za, without supposing it to imply a dread of water. That names with a meaning must often be used in a colourless manner may be seen not only by remembering the fact already noticed that normally all proper (or non-descriptive) names have once been general (or descriptive), but also by asking ourselves how else could either the application or the meaning of any general name ever be altered. If, for instance, certain chemical substances now supposed to be elements are ever found to be really compounds, we thereby discover that some (so- called) elementary bodies are not elementary; or ' REFERENCE-NAMES ' 1 83 if the epithet worthy changes its meaning, that points to the fact that some of the so-called worthy people are seen to be not perfectly worthy of admiration. Wherever, in short, an established descriptive name is discovered to be not quite rightly applied by custom, the proposition stating that discovery must be a contradiction in terms — and would be absurd in a more effectual sense if the mere words, instead of the speaker's meaning, <:ould make an assertion self-contradictory. It is only, however, by using the subject-term in a colourless manner that we can in these cases ■escape real self-contradiction. The use of names we are here considering is neither unfamiliar nor difficult to understand. Whenever a name is used as a mere subject-term — that is to say, not as indicating (through the con- notation of the name) what is denoted, but merely referring to it, as, for instance, the name ' Trafalgar Square ' refers to a certain part of London, then •objections to the use of the name become as trivial and irrelevant as the objection that the angles of Trafalgar Square are not exactly right angles, nor its sides exactly equal ; and along with the disarming of such objections the Socratic method of demand- ing exact definition loses all its point. So that if an assertor can make out that a vague name he 1 84 DISTINCTION uses is a ' reference-name ' of this description, he thereby puts forward a sufficient excuse for its vagueness. The question then arises, How is the assertor to support his claim that a word he uses is only a ' reference-name,' and so need not be defined ? The power of pleading this excuse is a sort of contro- versial advantage, and as such it must somehow be paid for. The assertor's right to waive so pressing a question can hardly depend on a process so easy as that of his merely calling the word a ' reference- name,' for we are not now laying down fanciful rules of debate, but are seeking to extend and generalise an already accepted practice — a practice which justifies itself to assertors and critics equally. There are various possible ways of describing the price that one must always pay for waiving a question, but they may be summed up in saying that one binds oneself to remain, for the occasion, wholly unprejudiced in regard to it. That is the pith of the matter, whatever kind of question it be that is waived. And in regard to the question how a term shall be defined, it is the ' meaning ' (the con- notation or associations) of the word which the assertor who calls it a reference-name declares his willingness to leave entirely out of account. He 'REFERENCE-NAMES' 185 thereby gives notice that he will not take any ad- vantage of its meaning, will not use it argumen- tatively, or beg any question by means of it. In discussing the question whether science is a true guide, for instance, we must not base our answer on the meaning of the word ' science ' ; and the same in discussing the question whether a general name can be ever non-descriptive. Let us see more widely, however, what this means in practice. It means, in the first place, that he who claims to use a name for reference only, thereby limits the purpose for which he uses it. He uses it, as the logicians say, to denote only — not to connote. The essence of proper names, as we saw above, consists in the fact that those who use them agree on their ' denotation,' on the individual things or cases that the name is taken to apply to ; the received application of the name ' Trafalgar Square ' is clear enough, although the Square may fail to satisfy the mathematician's ideal. And since the whole enquiry after exact definition is casuistical — that is to say, is aimed at finding the correct application ' of the name — such an enquiry naturally becomes purposeless if the application of the name is clear, or is taken for the ' Its actual, not merely its ideal, meaning. See pp. 22-28, and Chap. x. , i86 DISTINCTION moment as clear, already. The occupation of the critic is gone as regards that name ; only that of the pedant, or verbal purist, is relevant. And in the second place it means that the assertor is willing, if required, to substitute the contradictory of the name for the name itself, non- A for A, or A for non-A.' Nothing short of this willingness can satisfy the critic that the disclaimer is genuine. Of course, however, there must be no question-begging by means of the name, on either side ; and whatever name be used its denotation ^ must be agreed upon ; if our audience wish us to call the scientific men 'the unscientific' (or the religious * the irreligious '), we may willingly yield the point so long as they mean the same people as we do, and do not try to beg the question of fact by means of the name. We shall presently expand the meaning of these brief statements with the help of examples. Mean- while there is one further general question to raise. We have seen how a subject-term may become a reference-name ; and, if we assume for the moment that every term as used in an assertion is either a subject or a predicate,^ the question suggests itself ' Or, more simply, to ivithdraw the name. ^ See p. lOO, note, ' Strictly speaking, I consider this analysis insufficient for any 'REFERENCE-NAMES' 187 whether a predicate under any circumstances can become a reference-name. I wish to show that predicate-terms do essentially resemble reference- names in so far as they are, for a passing purpose, purely symbolic. The case here referred to can, perhaps, best be seen in its full generality under the notion of ad- missions that are made ' for the sake of argument' — an everyday practice enough. Perhaps the best- known occasion of such admission arises when we try to push someone else's assertion into absurd consequences, but this is not the only occasion when it is used. Any hypothesis (or guess, or suggestion, or theory) that is started, whether in science or in casual conversation, is capable of being put forward in a more or less tentative and provisional way ; and the hostile critic obtains for himself the greatest advantage if he can make it appear as put forward dogmatically. It is easy to recognise that human theory is fallible, so long as we do not undertake to prove it wholly untrue. Certain Roman Catholic opponents of Darwin's theory, for instance, cleverly take this line.' But thoroughgoing logic. But we may adopt it here as widely accepted and as perhaps sufficient for our immediate purpose. ' Besides Mr. St. G. Mivart, whose writings are widely known, the Rev. J. Gerard, of Stonyhurst, has published a most interesting series of articles proving that the natural selection theory, if treated i88 DISTINCTION just in the same way as an assertor has the right to waive the question how some reference-name shall be defined — that is to say, by reducing the claim that is made by it — so the propounder of a theory has the right to propound it, if he pleases, as entirely wanting in dogmatic force. He may say in effect, ' Let us assume it for the sake of argu- ment, and see what follows from it, and how it accords with the facts we can observe.' It would be intellectual slavery indeed if we were never allowed to raise a question till we had found its infallible answer. And the impersonal interest in Darwin's theory, or question, survives even the most successful attempts to show that some of his followers — or even possibly Darwin himself, when off his guard — have been too confident. None of us, not even the Roman Catholics, are always sceptical. A theory is in a certain sense an assumption. It begs the question in so far as it pretends to give a dogmatic answer. But an assumption may, of as a dogma, breaks down. But he wrongly assumes that it must be so treated. These articles are well worth reading, if only for the wealth of close first-hand observation of Nature which they contain. Most of them are reprinted in two books entitled Science and Scientists and Science or Romance, published by the Catholic Truth Society. The fact that Father Gerard misunderstands also some of the details of the theory, does not affect the more fundamental mis- understanding which his books are here used to illustrate. 'REFERENCE-NAMES' 189 course, be made by leave of the persons addressed, and there is no harm in begging a question when it is done openly. Even a concession would be called an assumption with equal right. If I concede to your request that the class to which Professor Huxley belongs shall be, called ' the unscientific,' or the class to which Mr. Gladstone belongs ' the irreligious,' there is just as much (and just as little) assumption in so doing as in following the ordinary practice of using the names. .And, simi- larly, to assume any theory purely for the sake of argument is the same as to concede it, for that purpose, to someone else. In the light of these truths we may now see the only manner in which a predicate-term can escape for the moment the demand for its defini- tion. The same idea is familiar enough to us in the shape of the recognition that words have a purely symbolic function. In mathematics, and outside mathematics, we are accustomed to manipulate untranslated symbols freely, leaving the question how to translate them until some other process has been performed. Symbols may even stand for absurdities and yet be used in a process of reaching intelligible truth. And precisely the same is true of every predicate-term so far as it is regarded for the moment as merely I90 DISTINCTION the middle term ' of an argument — that is to say, as needing some major premiss ^ to show what is really intended in using it as a predicate in the given case. If all assertion be analysed, as logic often analyses it, into terms and relations between those terms, then we soon find that enquiries as to the exact inter- pretation of given terms and given relations not only can be conducted in complete independence of each other, but are best so conducted. And this means that in interpreting any predication there is a certain part of the total process where it is a clear gain to neglect the enquiry as to the meaning of the terms and to concentrate attention on the meaning of the relation asserted.^ For this fleeting purpose, then, a predicate may essentially resemble a reference-name, in the fact that enquiries into its definition are irrelevant. But a difference so subtle as that between the dogmatic and the undogmatic use of names requires illustration in order that the use of the distinction may be fully seen. And some examples can be given in the course of considering generally the criticism of distinctions in refation to belief. To this, therefore, we may now pass on. ' See Appendix, p. 267. '^ Ibid. ' The old logical distinction between the formal and the material! validity of deductive reasoning was itself an early, if too special, recognition of this widely general truth. CHAPTER XVI THE CRITICISM OF DISTINCTIONS IN RELATION TO BELIEF The word logic is as hard to define as the word philosophy, and for the same reason : we are all of us always logical — imperfectly, and none of us ever logical^perfectly. Yet we mostly agree that, just as the criticism of assumptions is the specially philoso- phical function, so the criticism of beliefs (or ' judg- ments') is the specially logical one. And in adopting this view I make no pretence of drawing a line between assumption and belief; by assumptions I mean here only the deeper and more ingrained beliefs ; and if any reader requires it I must, therefore, concede that logic and philosophy are one. It is far from an easy matter to reduce to order, for purposes of general survey, the vast variety of occasions on which logic, so understood, comes into operation ; and especially if we admit, as I am willing to do, that in all actual thought, however apparently unchecked, there may be some 192 DISTINCTION conflict between the assertive and the critical spirit. Nevertheless, we can simplify the task by turn- ing attention to the openly controversial side of thought, where the struggle between assertion and doubt passes out of the unreflective or tacit stage and becomes conscious of itself. We may set up the aim at understanding the nature and rights of objection generally. This includes both the attack on beliefs and their defence, since we cannot defend an attacked belief without thereby objecting to the attack. When we object to our neighbour's opinion we sometimes take the cautious line of asking him for the grounds of it, or (what comes in the end to the same demand) asking him to define the mean- ing of his assertion clearly ; but often enough we risk a gratuitous opinion of our own as to the cause of his error. And in the long course of the history of logic a great many names of special fallacies have been invented, some of which — for example, 'begging the question' — may doubtless be useful on occasion. I do not, however, propose to offer any remarks on these, since the method of asking for proof (or for definition) is exhaustive, while the attempt to specify the error committed is not so ; and besides, it is much more easy to accuse your opponent of committing some special fallacy than to be perfectly sure that the accusa- THE CRITICISM OF DISTINCTIONS 193 tion is a just one. At any rate, it is only your opinion against his, and no general rules can decide between you. The special assertion or argument depends on the whole of its context. At the beginning of Chapter X. the assertion was made, in passing, that ' all criticism of judg- ment is criticism of distinction ; ' and we noticed there the reason why common-sense is not inclined to adopt this view. The criticism of distinctions in its most direct form, at least, is apt to seem rather wordy and frivolous when applied to beliefs about matters of fact, however well suited it may be to the discussion of ideals like justice or culture. If we want to dispute an assertion, for instance, that the prisoner did the deed, or that the moon has such and such an influence on the tides or on the weather, we generally do not find the terms of the judgment ambiguous, but we ask how the judgment as a whole accords with other facts. The moon, the tides, the weather — as a rule people agree sufficiently as to the facts these names refer to ; and a critic who took the line of disputing their interpretation would generally be thought to be quibbling, and quibbling weakly — trying to bolster up a desperate case. Two things should be noticed, however. One is, that we cannot pretend to press the distinction O 194 DISTINCTION between matters of fact and matters of theory or opinion ; we can only use it by leave, and for strictly limited purposes. And the other is, that even where we take the meaning of the terms of an assertion as agreed upon, yet the exact relation asserted between the terms is often conceived with vagueness. The notion tiiat A caused (or generally causes) B is by no means so simple as it seems on the surface. As will presently be shown, there is plenty of room in it for difficulties as to its precise interpretation. The bearing of the first of these two admissions on the criticism of beliefs may be shortly stated as follows : — Since questions of fact are in the end questions as to the exact nature of a fact, they are always questions as to whether a fact is correctly conceived or described. There is no dispute, for instance, that a ghostrseer or a person in delirium tremens has some sort of vision ; the only question is, whether what he ' sees ' is correctly conceived or described. So that, on philosophical grounds, it is better to recognise that all questions are really questions of theory or opinion, some being acci- dentally more easy to settle than others, and these being loosely called questions of fact. At any rate an asserted fact cannot, on the mere ground that it is ' fact ' and not ' theory,' escape criticism as THE CRITICISM OF DISTINCTIONS 195 regards the exact conception of it ; rather, it is where we agree to waive such criticism that we call it a ' fact.' And if all questions are really questions of opinion— «>. of the way in which some £aet is conceived— ^then ^11 objections whatever to any judgment, whether of fact or theory, may be viewed as complaining that the conception — ^and so the meaning of the assertion — is not clear. The truth or falsity of an assertion does not, then, attach to the words in which the assertion is ex- pressed, but to the meaning they are inteiided or taken to beaff. Just as every ' fact ' rsaXly is a fact of some sort, so every belief is an attempt to see a truth which J? really there to be seen if our minds could see it. The believer, as a rule, ' means not, but blunders round about a meaning ' ; ' and the important question is, how far his meaning, such as it is, falls short of the truth of the matter. So that to question the truth of an assertion is to ask the aggertor exactly what his meaning is, and to ask this is a way of qvfsstiontng the truth of what he asserts. The demand for strict definition of the terms is, therefore, only not pressed when the critic thinks he sees a worse ambiguity, a more important vagueness of conception, in the relation said to exist between the terms. And it is an accident — ' Pope : Epistle to Dr. Arbuthnott o 2 196 DISTINCTION that is to say, it depends on the controversial needs of the moment, as they appear to the critic — whether the attack upon the way a causal relation is conceived shall take the form of an enquiry after a dividing-line or not. It can always take this form, although, for the sake of rhetorical effect, some other form often seems better, as being more direct and less elementary. But even in this latter case it may be a gain to the critic to know the rights of the more elementary attack. Let us, therefore, ask in what sense it is true that all assertion of causation is liable to ambiguity. In order to make this clear we have only to remember that reasons for deciding in favour of any theory are always, when closely looked at, reasons for deciding against some one or more alternative theories ; just as reasons for deciding in favour of any one predicate are reasons for deciding against some one or more alternative predicates : so that in either case the question what exactly is the assertion is of extreme importance before the question as to proof takes definite shape, and is thus bound up with the question as to proof indis- solubly. For instance, the modes in which two ' things ' (or qualities or events) may possibly be related to each other in causation are so various and subtle that the value of any such assertion. THE CRITICISM OF- DISTINCTIONS 197 where the things related are unambiguously de- scribed, mainly depends upon the fulness of detail with which in the special case the relation is conceived. An assertion that goes no further than that the two things in question are somehow •connected, either contains beneath its surface a hint as to the kind of connection supposed to •exist, or else it tells us nothing at all. In a uni- verse such as science supposes ours to be, all actual events are somehow connected ; so that the origin of any ' thing ' and the origin of any quality of that thing are (since all origins are events) some- how linked to that of anything else we choose to mention. It is a truism, therefore, to say that A and B are ' somehow connected,' if this be literally meant ; the only problem of theoretical and practical interest is as to the exact nature of their connection. Is the dependence (or repulsion) between them mutual or one-sided ? Is it direct or indirect ; and, if indirect, what are the steps left out? Under what conditions may the presence or absence of one be regarded as a sign of the presence or absence of the other .' These and similar questions, and the answers we give to them or suggest, are what fill out the meaning and create the value of our otherwise empty assertion that A and B are ' connected.' 198 DISTINCTION But, however we picture the connection in the special case, we are apt to do some violence, in idea,, to the fluent transitions of Nature. If Nature is,, as we all in our feasonirig moods admit, a seamless fabric — a process that never really makes a leap, though it often seems to do so, then there is no such thing as ' direct ' causation between any things or events as named ; there are always inter- mediate steps that either our clumsy observation overlooks or our practical aims forbid us to notice. So that wherever we claim to see direct connection a distinction is implied where no real gap exists — a distinction which is artificial in the sense that man's vision, limited either against or by his will, has drawn it in seamless Nature.' All judgment, all belief, thus bases itself upon alternatives that are used as such. Whether we regard assertion as predication of an attribute or as having reference to causation, the dependence upon alternatives, for meaning, is the same. The reduction of all judgment to predication is very familiar to readers of logical text-books, and when ' An assertion, e.g., that X is directly the cause of Y, loses its value if X is admittedly only related to Y as the bud to the flower : in poetry perhaps we can speak of the child being father to the man, but science rather regards childhood and manhood as roughly- marked stages in a course of development of which we have not yet understood the really initial impulse. THE CRITICISM OF DISTINCTIONS 199 it is so reduced it is easy to see (as we saw in Chapter IX.)that the meaning of the predicate-term depends on our taking that term, and its contra- dictory, as sharp-cut alternatives. And the reduc- tion of all judgment to judgment about causation is equally sound and even more useful. To pre- dicate is to say something about the ' nature ' of the subject ; but the nature of any subject consists in its relations to other things — its origin, and the way it behaves to its surroundings. If we call a man a hero or a rascal, we mean that he may be expected to behave in certain ways. The full meaning, therefore, of any predicate can only be seen in the light of full knowledge of causation, and the closer we press enquiry into its meaning, the more we are thrown back upon causal enquiry. Predicates are thus only shorthand registers of causal assertion. It will possibly help to clear up some almost unavoidable difficulties of expressions in what has been said above, if we notice that unreal distinctness may always be viewed as the neglect of a midway alternative. If A and non-A are only roughly distinct, that means that between them lies the third possibility, ' both at once.' The use of this view of the matter is in connection with the attack on judgments generally. In all judgments as to 200 DISTINCTION the nature or the cause of a fact we proceed to a more or less firm and final conclusion by means of a gradual weeding out of faulty alternative theories. After one or more possible alternatives are dis- carded we conclude that what remains is the essence or the cause ; only, we often have a too limited vision of what remains. And one of the commonest causes of such limited vision is the clear and sharp separation of alternatives that in reality overlap, thus forming a midway alternative between the two extremes ; all judgments based upon the distinction are liable to be correspondingly incomplete. If we admit that all judgment thus lies open to an attack the result of which may be foretold by anyone who sees that Nature is continuous, we are getting very near universal scepticism. And, indeed, one main purpose of our enquiry was to raise at the end the question how far reasoned doubt must triumph over certainty, and whether a casuistic treatment of definitions need or need not eat the heart out of the faiths we live by. The next chapter attempts to discuss this question in the light of the notion of 'reference-names.' Meanwhile it only remains to notice here that logical criticism depends on two suppositions — that a definite meaning is intended by the assertor, THE CRITICISM OF DISTINCTIONS 201 and that what is said is put forward as reason- able. Logical criticism is at once disarmed if the assertor refuses to allow the critic to make these two assumptions ; that is to say, if the assertor admits that he is only talking vaguely, or that •what he asserts is not a reasoned judgment. But in practice assertors are slow to make such admissions, unless they are driven to it. As a rule they preserve as long as they can the appearance •of knowing clearly what their own assertion means, and of having some intelligible ground for believing it ; which is only natural — for although by admit- ting that your assertion is meaningless or ground- less you take the wind out of the critic's sails, you •do not thereby convince the audience that your assertion is true. CHAPTER XVII SCEPTICISM AND CONCILIATION By ' scepticism ' is not here meant any merely theological unbelief, but something more wide reaching ; a mental state which certainly includes theological unbelief, but which brings with it far more danger to mind and character. For there have been notable exceptions to the rule (if it be a rule) that heretics and ' unbelievers ' are lax, or deficient in moral fibre or in mental receptivity and effective- ness, and even among our less notable friends some of us may know unorthodox people whose charac- ters are strong and healthy and whose minds are clear. It lies out of our way, at any rate, to raise any questions here about theological, as distinct from general, scepticism. And scepticism in the sense in which some ot its opponents try to define the term lies equally out of our way. A scepticism which can erect itself into a system is already something other than sceptical ; it is too self-satisfied to deserve the SCEPTICISM AND CONCILIATION 203 name it assumes. We are none of us always and only sceptical. That is why there is an air of the pl&yground about the proposal that scepticism pure and simple should be adopted as a philosophy. The proposal so made is felt to be unpractical and only half-serious, a debating-society question, useful at most for the exercise of thought it affords. Sooner or later the artificial sceptic of this kind lays himself open to a tu quoque retort ; and hence the recognised answer to any of the philosophical Systems of scepticism is homoeopathic, and consists in showing that they are not so sceptical, after all, as their authors fancy they are. They pull down certain assumptions indeed, but only by the help of other assuittptiofls which are quite as hard to justify. It would be very convenient for the opponents of scepticism if these were the only sceptics that had to be taken into account. As things are, the doctrine that scepticism 'casts itself out ' is not so destructive to scepticism as it at first appears. The genuine sceptic may claim with equal right that the supposed remedy is more sceptical than it pretends to be ; that it supports his idea, but only destroys his imperfect realisation of it. Scepticism in its most complete and genuine form is the spirit that questions, not the spirit that denies. For immediate practical purposes it may 204 DISTINCTION be much the same whether we doubt a belief or deny it, but there is a great difference in the con- troversial strength of the two positions, and conse- quently in the ease with which a remedy can be found. He who denies may be met with the very same weapons that the doubter uses against asser- tion generally : there is far more kinship between assertion and denial than between either of them and doubt. Doubt is a more pressing trouble nowadays than denial, and seems to be part of the price we are paying for the recent advances of science. How far the mental state may be alsp clue to bodily causes I do not here enquire, nor whether a mental sickness hangs over our times, as over certain former times, like some mys- terious epidemic. Assertions of this sort are easy to make or deny, but hard to prove or disprove. And though much of the malady be laid to the account of causes such as these, external to the mind of the person affected, yet it seems reasonable to hope that something may also be done towards finding a remedy by way of an appeal to the scep- tical mind itself As we have now sufficiently noticed, the special feature or symptom of the disease in question is casuistry — by which is meant not only moral casuistry, not only the raising of difficult questions SCEPTICISM AND CONCILIATION 205 of detail in regard to moral rules, but the tendency to compare general laws and general notions with concrete cases wherever the former are found — an unquiet spirit that is always asking where the line shall be drawn, and whether this or that case comes under a given name. Moral ciasuistry is, no doubt, the most familiar work of scepticism ; our reasons for conduct are a favourite target for the spirit that doubts. Why should we do this or that disagree- able duty? Perhaps we are foolishly ascetic in supposing it a duty at all — are cutting ourselves off from innocent amusements just as the people do who think it wrong to whistle a tune on Sunday. Follow the voice of conscience ? ' Ah ! if only I could learn to hear it unmistakably. But I find that other voices always mingle with it — the voice of pride, for instance, and that of fear, and of many other faulty human qualities that belong to my nature and will never be rooted out. Give me more light, if you can ; but do not tell me to follow what seems the voice of conscience, for I am not wholly divine.' In some such way doubt attacks at times even the best of our moral rules, but its influence extends also beyond morality. The notion that Nature is continuous and that the lines that language draws are artificial, has certainly taken root more strongly in the minds of the 3o6 DISTINCTION present generation than ever before, and hy its aid the answer ' yes ' and the answer ' no ' to any ques- tion can be drawn so close together that each of them fails to get a sharp background of meaning for itself. Much of the virility of our beliefs may thus be lost, whatever their subject-matter. When our ' yes ' and our ' no ' are one, it ceases to make ?iny difference which w^ answer. Although, as w^ saw above,' the call for a de- finition is in the end the sgjne as the caJl for grounds of belief, yet the spirit of doubt sometimes prefers to show itself in the latter form- It is often easy, in this form also, to raise questions that pannot be answered finally, or can only be answered by the brute force of assertion. Is there not, at least, ' room for error ' in this or that belief, the sceptical mind enquires ? is there anything — anything what- ever—about which the individual is not liabk to be deceived } Is any ' fact ' pure fact ? does not all observation contain an admixture of human and faJilible theory ? Can we get outside ourselves and see things as they really are, instead of as we are compelled to see them through the spectacles which nature or custom has provided — nature whose instruments are all progressive and improv- able ; custom whose power is local, and subject to Chapters x. and xvi. SCEPTICISM AND CONCILIATION 207 hunian caprice ? Are not truth and falsity ideals ; who can say that they are ever perfectly realj^^ed in this or that actual belief? Is there any lipiit, other than a merely practical one — if even that^ to the wisdom of reserving our judgment on a given point? Put forward thus, undogmatieally, in the form of a question, such doubts are hard to cojaquer. The old and easy plan of arguing thai if nothing can be known for co-tain, then it cannot b? known for certain that nothing can be known, falls here beside the mark. The sceptic of the more njodern type does not assert that nothing can be known ; he is content to ask what is known. Help me to find a single instance, he pleads, of an undeniable truth which is not merely a truism. Most of our gup- posed Jaws, or uniformities in Nature, are admittedly only approxipfiate ; but even pf the Iea§t deniable of them, such a$ the law of gravitation, can we say more than ^at they are proved by experience, and that a wider experience may show theni to be of only limited truth and value ? The sceptic's busi- ness is not to deny accepted truths, but to seek for their explanation, and so to secure them, if possible, against the charge of being 'merely empirical.' But he cannot be content with the false security that comes of reducing the truth to a truism, nor 2o8 DISTINCTION yet with reliance upon the help of notions — such as causation — ^which themselves get their security ^ in the same manner. State the law of causation in any way that is not truistic, and the sceptic is ready to show us how insecure it is. Frame any natural law without reference to causation, and he is ready to show us that it is merely empirical. By an ' empirical ' generalisation he means one that rests on mere number of observations. Where a generalisation is thus supported it remains pos- sible that all the cases observed are under some limitation or condition which makes an essential difference. However many samples of sea-water I taste I shall never justify the induction that ' all water ' is salt. It is a truism to say that we must multiply observations ' as far as the case requires ' ; this, no doubt, the often-quoted king of Siam thought he had done when his experience led him to conclude that water could never freeze. How, then, can we, the sceptic asks, make our ' truths ' secure on all sides against reversal ? We are constantly finding that our views of the truth have been too simple, too abstract ; that something which we have believed to be true absolutely, or on all occasions, is only conditionally true — true on a certain set of occasions, or true when certain ' See p. 127. SCEPTICISM AND CONCILIATION 209 circumstances, apparently insignificant but really essential, are present ; we are constantly finding that owing to some detail that has been overlooked, some unwarranted assumption that has'been made, our answer ' yes ' to some question requires to be altered into the opposite answer, ' no.' It is always easy to raise the sceptical question. Where, then, is the process to stop ? To anyone who adopts in the main the views suggested in the preceding chapters, it must seem likely that the next movement in philosophy will pay more attention than has hitherto been paid to the shortcomings of language as an instrument of thought. Even our most philosophical questions must, after all, be stated in language, if they are to raise any interest, or serve any purpose, outside the solitary and dreaming mind ; indeed, they can hardly be called questions, even to the solitary seer, so long as their answers escape formulation and appear only as passing glimpses of hardly com- prehensible truth. Few must be the philosophers who are wholly unaware of the danger of encourag- ing mystical insight, and when they forget it for a moment the rest of the world are ready enough to apply the common-sense remedy. Wherever lan- guage is used, it is to the intelligible conditions of the use of language that we must make appeal. P 210 DISTINCTION Of course it is easy to say that question-raising^ must stop at the point where questions become unintelligible. The difficulty is to find that point, or at least to find it to our neighbours' satisfaction as well as our own. Our neighbour who uses a phrase that we find unintelligible may rightly think that the fault lies with us and not with the phrase ; or we, who discover an ambiguity that his clumsier vision overlooks, may be right in calling attention to it. Some people find it impossible, for instance, to answer the question whether ' Nature ' can be spoken of as something real, or whether there is such a thing as the ' Luck ' that gamblers believe in, until the terms of the question have been purged from ambiguity ; but anyone who confidently answers 'yes' (or 'no') to such a question will natu- rally think them only quibblers. Which party is in the right ? If we were only infallible, or if only we had an infallible authority to appeal to, how convenient it would be ! But the old centres of authority notori- ously do not satisfy everyone, and when we attempt to set up new ones the disputability of the claim is even more likely to be forced upon our notice. As things are, it seems that we can only choose between two courses with any hope of satisfaction : we may either assume that, though not exactly infallible, SCEPTICISM AND CONCILIATION 211 we ourselves are as a matter of accident always right and our neighbours always wrong, or we may seek for agreement somewhere with our neighbours, using that as a point of departure ' for the sake of argument.' It is this latter course that I wish here to explain and defend. The method is simple enough, though not yet very familiar. In one important respect it is the opposite of the controversial method which we all know so well. It aims at conciliation ; it proceeds by making the best of our opponent's case, instead of taking him at his worst.' Already there are signs of increased readiness to try this plan. It is felt, and especially by those who are most experi- enced in the possibilities of controversy, that the old game of ' idealisation and caricature ' is one that both sides can play at for ever ; that the prac- tice of scolding and calling names, politely or otherwise, leads to no conviction ; that the most interesting part of every disputed question only begins to appear when the rival ideals admit each other's right to exist. Always the practical ' I admit, of course, that the ideal of conciliation, like all ideals, requires to be judiciously combined with the opposite ideal, in practice. Still, as against the controversial tone which is still most commonly taken, it may have some value ; as against, for instance, the tone taken by each party in the dispute about Culture, above referred to. P3 212 DISTINCTION problem is how to enable each ideal to limit the other. With the concessional or conciliatory method as a whole we are not here concerned, but only with its bearing upon the rivalry between assertion and doubt. In regard to this the proposal is that, instead of separating, as at present, into opposite camps and waving our rival banners, instead of doing our best to magnify our difference from our opponents and so to win a possible party advan- tage, we should seek first the common element from which neither side will ever be really free. None of us are merely sceptical, none of us quite unshaken by doubts ; assertion and doubt are only abstractions, and so let them fight their battle within ourselves with the help of our neighbour's somewhat different experience of the contest. Why should we be in so great a hurry to label ourselves, to profess our faith or un faith? Who can say for certain what he believes, and with how much steadfastness ? Who is there whose belief- — as distinct from his creed — remains quite the same even from one year to another, on any subject except those where there is least interest or least difference of real opinion .' Wherever insight is most wanted, there clouds and false lights most obscure and delude our vision. SCEPTICISM AND CONCILIATION 213 Though separable by abstraction, yet in all actual thought doubt belongs to assertion in a very intimate manner. If there were no room for doubt there would be none for assertion— a fact which explains the effectiveness of epigrammatic paradox.' In an important sense, every assertion gets its meaning — its assertive force — through its doubt- fulness. That is to say, unless it answers a ques- tion — unless it chooses between 'yes' and ' no' where either answer is regarded as possible — the assertion is truistic, and so not properly an assertion at all ; it tells us nothing that we did not know before the assertion was made ; the question answered is only a question begged. For instance, Is a straight line the shortest distance between its extremes ? If this is to be more than a verbal definition, the term straight line must be so defined ^ as to avoid begging the question raised. Accordingly, there can be no such thing as a truth (an applicable truth as distinct from a truism) that is recognised as ' necessary,' in any other sense than that it is in fact accepted as true. The distinction between proved and intuitive truths, regarded as a distinc- tion applicable to actual cases of belief, melts into ' E.g. the statement that 'Genius is an infinite capacity for taking pains.' ' E.g. by the movement of a point. 214 DISTINCTION that between cases where there is respectively more and less reflection upon the grounds of belief. Wherever a belief is more than merely verbal, the question what determines our choice between the ' yes ' and the ' no ' is a possible question, and cannot be answered by reference to our own postur lates as to the meaning of the terms employed. However diiificult it may be in a given case to dis- cover the source of our certainty, the search for it can never therefore be wholly without importance. And the search for grounds of belief is exactly the sceptical function. If this be a sufficient excuse for our human tendency to doubt from time to time, the opposite tendency — belief and assertion — is even more easy to excuse. The reasons against hesitation are, in fact, much more widely recognised and remembered than those in favour of it, since it is obvious that all action implies the absence of hesitation, for the moment, and that he who hesitates is often lost. And besides, faith — even faith against reason — has been preached and practised so long and so well that all language is full of words that get their meaning from this ideal. We are none of us merely sceptics. It is even said, and perhaps with truth, that unbelievers in theology are apt to be overcredulous in other SCEPTICISM AND CONCILIATION 215 ways, as if the inclination to believe without grounds required an easy and familiar outlet, in order to render it harmless. At any rate, the fact remains that in every human being the critical faculty does lie dormant during part of the time he spends in living. And the question suggests itself. Cannot we use this fact in finding a solu- tion for the general problem as to the limits of scepticism. With the help of the notion of 'reference- names ' I thipk we can. Wherever we take a word or distinction at secondhand, without enquiry and without ourselves laying stress on it, there we are using a 'reference-name.' Suppose, for example, the question is raised whether we know reality or only appearance. Opponents of scepticism who (for the moment at least) belong to the old school of controversialists, picture the sceptic as himself laying stress on the distinction between reality and appearance, in the act of asking the question.' But we who are in no way concerned to regard the sceptic as foolish are not compelled to make so bold an assumption about the state of his mind. Perhaps he is only taking the language of his question at secondhand, and ' See, foi instance. Professor Caird, in his book on The Philo- sophy of Kant, p. IS (1st edition). 216 DISTINCTION using it as he supposes his neighbours use it and subject to their approval. At any rate, until there be some agreement between the parties as to the meaning of the distinction, the question cannot be answered ' yes ' or ' no,' for it cannot be asked with a meaning. And instead of the deadlock counting against the sceptic, it counts to this extent in his favour — that the doubt whether there is any (actual) reality as distinct from appearance is the very doubt he was trying to suggest. 'The sceptic,' says Professor Caird, ' has first to justify the division of the terms ere he can make it the ground of his scepticism.' If the sceptic in question be also. a controversialist, he is likely to answer, ' How very convenient for the other side ! Those who are not sceptics may insist, it seems, on the validity of any distinction that pleases them, and before the sceptic may say a word against their procedure he must first 'justify' what they have done!. ' Abolish the distinction, by all means, if you like : I was only trying to insist upon its faults.' ^ Similarly, we cannot decide the question whether there is anything supernatural by saying that unless an affirmative answer is given to it the predicate ' supernatural ' would have no meaning. ' See also an example given at p. l6l. SCEPTICISM AND CONCILIATION 217 This would be to go back to Cartesian ways of thought.! It is never necessary to suppose that a distinc- tion used by our opponent is used argumentatively, and least of all where his object is to break the distinction down. On the contrary, we are more likely to do full justice to his thought when we neglect to take the obvious controversial advantage which his words — used really as a concession on his part, not an assumption — may appear to give us. Here, for instance, lies the weakness of the argument ^ that A can never be developed out of non-A. He who suggests the possible develop- ment of one from the other is 1-eally questioning the distinction, not laying stress on it ; and, there- fore, to make it binding upon him is to beg the question he raises. To make names overrule facts in this way — to prevent our asking whether so and so has had such and such an origin, on the ground that our existing language assumes a firm line of division', is to suppose that in using lan- guage we are compelled to adopt the assumptions made by those who used it first. Against this natural superstition the notion that words may be ' See Appendix, p. 264. ' This argument is a familiar one in philosophy. An instance of it was given at p, 20. 2i8 DISTINCTION used 'for reference' is of some value. To ask whether non-A contains the germ of A includes the question whether our received distinction between A and non-A is not, after all, faulty. The questioner uses these names merely because he supposes that his audience will use them too. If not, so much the better for his argument. To preserve the distinction is no concern of his ; rather, his very object is to break it down. When the plan of conceding, instead of assijm- ing, is carried as far as it can be, the assertor and the critic reach a state of agreement which claims to be no more than provisionaj, but which may in given cases never be disturbed. The ' truth ' they agree upon is recognised by both as questionable, whenever they care to question it, but in the meantime as useful for the only purpose for which truth is ever useful — that of forming a step to other (questionable) truths as yet unseen. What more can we ever do with a truth than use it ' for the sake of argument ' } And this we can do as well with a false ' truth ' as with a true one, the difference being that false belief leads sooner or later to absurdity, and so to the correc- tion of the parent error, while true. belief is justified by its fruits; The plan of conceding instead of assuming is SCEPTICISM AND CONCILIATION 219 capable of application wherever a difference of opinion exists. The essence of it lies in dis- claiming any but our opponents' definitions, and so taking all our words as far as possible in the sense in which our neighbours understand them. We have seen how it is that the sceptical demand for a defini- tion becomes on occasion irrelevant ; it is out of place wherever the assertor is professedly taking his language at secondhand and using it only as he finds it used in some argument to which his own remarks in the end refer. And in view of the fact that even apparently independent assertions do not stand wholly out of relation to the reasoned views of other people, may we not apply this method wherever an assertion comes before us as doubtful } There are some who seem to fear that philosophy is degraded if taken as at all concerned with the clash of mere opinions entertained by different individual minds ; and of course we may all admit that absolute truth is a very desirable object. But it surely needs some special training or bent of mind before this or that doctrine can be taken as satisfying the ideal ? It seems, rather, that until some authority is equally accepted by all minds, every opinion that can bear to be stated in language does come before us (when we reflect upon it) as an argument and not as a revelation ; and 220 DISTINCTION that while this is so — until the new master appears who will win universal mental obedience in place of the limited following which every master has hitherto had — opinion of all kinds must be taken by human minds as true or false not in any final sense, but by comparison with whatever rival opinions are in the field. For this purpose we certainly need the power of using language purely as we find it used by others, and of declining to supply those others with a satisfactory meaning for their own expressions. If so, we must allow the ' sceptic ' the right to make an assertion ad hominem, an assertion only intended to convey a niedning to those who already make the assump- tion upon which its meaning rests. Much un- convincing philosophical controversy might be prevented if those who possess what they consider a good system of metaphysics could bring them- selves to admit that not the only alteirnative is to believe in a bad one. That all who do not use the one best system use some worse one, maybe admitted freely. But we can use an instrument without believing in its absolute perfection — and even in order to show, or to discover, the limits of its imperfect usefulness. That all human thought is dominated, and its capabilities limited, by the general notions we at SCEPTICISM AND CONCILIATION 231 any time happen to possess, is a truth with far- reaching consequences. On the face of it the objection may be very plausibly raised that the full meaning of any such truth could only be apprehended by a superhuman intellect ; that to understand how far all human thought is essentially limited, or liable to error, demands from us the power of rising above humanity and thence taking our observations — a feat which, as the history of metaphysics abundantly shows, may possibly be performed so as to satisfy ourselves, but hardly so as to stand against the criticism of even our next successors. Meanwhile, however, a part of this comprehensive truth (if it be a truth) lies open for all to observe and to find instructive. With merely human vision we can see some of the differ- ence that exists between the notional range of different people or different ages ; can trace some of the causes and effects of such difference ; and thence can draw certain lessons respecting the probable relation of our own beliefs to those of minds with a wider and narrower range respectively. And the result of the process should be to avoid the supposition that any existing philosophy is final, and to obtain suggestions as to the direction in which to look for means of enlarging our present mental horizon. Philosophy is doubt, just as 222 DISTINCTION science is knowledge ; but as the problem for the latter is how to be knowing without becoming dogmatic, so the problem for the former is how to be critical without becoming hopeless. Our knowledge, whatever be its value, has one unchanging element in it and one that changes slowly. Our abstract or ideal distinctions are the same now as they ever were ; it is our application of them to actual cases that suffers a gradual change. So far as the former element alone is concerned, the question of validity seems to be irrelevant. The purely ideal distinction between truth and error, for instance, can never be assailed, since its validity is assumed in supposing any question intelligible ; if there were no such dis- tinction there would be no meaning in answering any question, nor, therefore, in asking it. But as regards the actual discrimination of truth from falsity, it is surely the case that our knowledge is progressive, and that to deny this would be to claim infallible wisdom } When, therefore, we are asked whether we are sceptics, it is difficult to answer the question so as not to mislead the questioner. As a rule, however, nothing makes the questioner more impatient than SCEPTICISM AND CONCILIATION- 223 to tell him this ; and, besides, the same truth can be conveyed to him in a less offensive manner by raising difficult questions as to the meaning and" force of dotsbt itself. If Nature is continuous, perplexities as to the exact meafling of an asser- tion can (as we have seen) always be found by looking closely and by refusing to grant assump- tions without proof. Meaning, therefore, depends upon a sort of agreement not to raise these critical questions ; depends upon a provisional compromise made — though tacitly — by assertor and critic together, for the sake of practical purposes. If all distinctions have their purpose and are valid only for that purpose, what (it may be asked) is the force, or even the meaning, of criticising any distinction — and therefore of casuistry, and there- fore of doubt itself .' It cannot consist merely in the suggestion that the distinction is artificial — for that fact is admitted already. It can only, in the end, be resolved into an appeal by the critic to the assertor, to revise and perhaps to cancel for a time the agreement under which the distinction is taken as valid ; and such an appeal is never likely to be successful unless it is based on reasons. Doubt, therefore, which springs from a spirit of mischief merely — which is aimed at random instead 224 DISTINCTION of being inspired by a glimpse of unfamiliar truth — is not a very terrible affair ; while the doubt which is serious, and is so inspired, is surely rather a friend than an enemy to those who remember that there is still some truth, on any subject, fof fallible men to learn. CHAPTER XVIII GENERAL SUMMARY The main purpose of this book has been to discuss the best way of dealing with ambiguity. Effective ambiguity, we found, has its source in rough dis- tinction ; and so our purpose resolved itself into that of suggesting an improvement on the common- sense method of using rough distinctions — a reasoned discrimination in place of a haphazard tact. - In order to make the suggestion intelligible several explanations were required. Chiefly, it was necessary to show the nature of this ' common- sense tact,' and the manner of its working, and to prevent some possible misconceptions as to the relation assumed to exist between common-sense and philosophy As regards the latter subject, the view here taken throughout has been, that though the dis- tinction between philosophy and common-sense is only one of degree, so that neither can be strictly defined against the other except as ideals, yet the 226 DISTINCTION two opposed methods may (since methods are always attempts to follow ideals) be sharply con- trasted—as the method of careful attention to details, or interest in exceptional cases, and the method of taking short cuts, or believing in general rules. As much as this was said in Chapter III., leaving over till Chapters XI.-XVII. the more de- tailed account of what this difference involves in regard to the special question before us. But in order to deal with the former subject — in order to show the operation of common-sense tact in using rough distinctions — we had to discuss the nature of rough distinction in the light of examples, and were led to enquire as to the amount of rough distinction that exists in language gener- ally. The discussion of these subjects was needed also for the sake of certain incidental aims of the book, which will be noticed presently. As bearing directly on the question as to the nature and the amount of rough distinction, the chief points of interest are these : — First,' a rough distinction is any distinction where there is a borderland of doubtful cases between the things which are distinguishable as A and non-A (or A and B) respectively. The phrase ' unreal distinctness ' is useful for describing the ' Chap, ii., p. 1$. GENERAL SUMMARY 227 fault which a rough distinction (or indefinite name) suffers from ; but since the question ' What is reality?' is one of the unsettled questions of philosophy, great care must be exercised in using this name so as not to involve ourselves in needless metaphysical difficulties. We therefore elected ' to mean by unreal distinctness only distinctness which is admitted to be unreal. And everyone admits distinctness to be unreal just wherever he recognises the continuity of Nature. The admission, however, of the unreality (or artificiality) of a distinction is always complicated by the fact that a self-contradictory ^ state of mind is implied by it ; distinction as such — distinction at all — is the separation of kinds, and the notion of separate kinds is unavoidably opposed to the notion of differences which are merely of degree. Some reasons for this unavoidable opposition are noticed in Chapter XIII.» So the question as to the amount of rough distinction that exists resolved itself into the question how far the continuity of Nature must be recognised. And here we found* that, strictly ' Chap, v., p. S9- * i.e. a changing state of mind, a vacillation between contra- dictories. » Pp. 162-4. * Chap, vi., pp. 71-79' a 2 228 DISTINCTION speaking, the conception of Nature as at all dis- continuous is an impossible one to justify ; that so far as, it is used at all it is artificial and incomplete. And if it be so, all distinctions are rough when closely regarded. This strict view, however, is apt to seem ' theo- retical ' — that is to say, unpractical. For practical purposes we are often content not to look so closely. There are certainly many cases where in practice we never think of Nature as continuous, and some cases where we should find it very difificult to do so, and quite impossible to do more than barely imagine the continuity — impossible to verify it. And in ordpr not to rest a strong case on weak supports, we turned attention away from the more fanciful cases and raised the question how far unreal distinctness (or the continuity of Nature) has any practical effect in confusing our judgments. As to this, two things seemed clear. First, that although the claim made by common-sense tact cannot be disproved (however much we may suspect its weakness) where all common-sense is agreed, yet wherever common-sense is divided against itself we get a case where the vaunted tact has actually failed. Of two opponents, both may be mistaken, but both cannot be entirely right GENERAL SUMMARY 229 — both dannot have put exactly the right discount upon the distinction. It is, therefore, to controversy that we must look in order to discover the range of effective ambiguity, and especially to those disputed questions that recur the oftenest and are least easily settled, and so are the most contro- versial. Secondly, when we ask what disputes are the most controversial, so as to find the strongest examples of the effect of unreal distinctness, we see ' that ' immaterial things '^abstractions or ideals — are the chief source of lasting and recurring controversy. The problem how to apply abstract notions in concrete cases is the essence of all the real difficulty that arises about distinction and definition, and the difficulty is artificially increased by our human tendency towards ' idealisation and caricature,' which again depends upon the- real difficulty which it increases. Having thus obtained some view of the extent, theoretical and practical, of the influence of unreal distinctness, we next (in Chapters IX. and X.) discussed the actual harm that unreal distinctness, and the consequent ambiguity, brings about. This divides into two enquiries, since the harm caused by unreal distinctness when the unreality is recog- ' Chaps, iv. and vii., pp. 55, 85. no DISTINCTION nised is different from the harm caused by unseen ambiguity. But as regards the latter it seemed unnecessary to do more than refer very briefly to what had been said in our remarks on rival ideals (Chapter IV.) and on the spoiling of words (Chapter VIII.). A name which is ambiguous — i.e. clumsy — binds together as ' essentially the same ' things which are in fact essentially different. A general assertion about a class named A may (if A is a clumsy name) only hold true of a part of that class — the A's which are A in the deepest sense of the word, for instance ; and then the broad assertion that all A's are so and so will be untrue and misleading Or, again, it may only hold true of things which are A superficially, and so may be misleading to those who take the word in a deeper sense. And, simi- larly, the predicate A will mislead us wherever the real difference between S * and the other members of the class A is more important than the resem- blance ; S, that is, may be A in a shallow sense only, and in a deeper sense may belong rather to the non-A's. It was, therefore, chiefly the effects oi recognising ambiguity that we were concerned to notice in Chapters IX. and X. The recognition of unreal ' The 'subject' ; that which is spoken about. GENERAL SUMMARY 23 r distinctness anywhere is the breaking down of a corresponding distinction. But meaning, like value, implies a comparison — a background ; the meaning of any predicate name — and equally of the predica- tion made by means of it — depends on clearness of distinction between that name and its contradictory. So that the meaning of any predication ' is lost where the predicate name is seen to be unreally distinct. The reason of this, we found, lies in the need for clear difference between assent and denial. If an assertion is to have any meaning, the answer ' yes ' and the answer ' no ' to the question it raises cannot mean the same — a reflection obvious enough to have been made long ago, and in fact to have been used as one of the foundations of the ' Socratic method.' But since (if Nature be continuous) all distinc- tions are, strictly speaking, rough, this seemed at first to prove too much. Surely not all predications are meaningless ? And so the question could not be avoided, whether we should seek to enforce this hard doctrine in all its strictness, or whether any compromise could be effected between it and the common-sense view that total absence of meaning ' And always the loss of meaning in a predication involves a corresponding loss of meaning in all general assertions to which that predication serves as minor premiss. See Appendix, p. 267. 232 DISTINCTION in an assertion is exceedingly rare. Does meaning stand and fall with absolutely sharp distinction, or does it not? We found, at any rate, that meaning does not depend in the least upon extent of borderland. A predication does not get more or less of meaning according to the less or more of borderland between the predicate and its contradictory. What it gets in that way is merely the extent of its liability to become ambiguous. And just as the actual mis- leading influence of a word — its actual ambiguity — depends upon other conditions than its mere lia- bility to mislead, so does the actual meaning of a word depend upon other conditions than its liability to be meaningless. It is true that a word which is ever so slightly indefinite is liable to be ambigu- ous, and so liable (for the time) to be entirely devoid of meaning ; but it is not true that it must be ambiguous — it is not true that the vaguest word in the language is unable to be used with a meaning which, for the time, is perfectly unmis- taken. We saw that from this it follows that though meaning does stand and fall with absolutely sharp distinction, yet artificial sharpness will make a meaning just as well as real sharpness, so long as the parties agree to accept its reality as (for the time) unquestioned. Meaning thus depends, in the GENERAL SUMMARY 233 end, upon agreement. Every meaning, like every misunderstanding, takes two to make it. Faith is required in order to understand. No general name can ever be used as a predi- cate without a certain amount of charitable inter- pretation. It is of the essence of the general name that it shall blur individual differences which are nevertheless real, shall bind together as ' practically the same ' things which we admit to be really different. And this is true whether by 'prac- tically ' be meant ' to all intents and purposes,' or only to some one intent and purpose that we try to keep in view. In either case a choice is made among various differences that compete for notice ; some we select as important, others as the reverse. And since the soundness of our choice is always (in theory) questionable, a certain mutual consent is required in order that any predicate may perform its function properly. In Chapters XI. — XIII. inclusive we discussed the improvement which the philosophical method should help us to graft upon our hastier methods of dealing with ambiguity. That improvement •consists, we saw, in regulating common-sense tact by making it conscious of reasons for laying more and less stress upon a distinction at different times. The central idea of such regulation is the attempt 234 DISTINCTION to take into account the special occasion of the use of a word, instead of judging of its ' sufficient distinctness ' broadly by lumping together, as if they were one, the whole set of various possible occasions. As said before, it is ambiguity itself, and not mere liability to it, that we are here con- cerned with ; and actual ambiguity attaches rather to assertions than to isolated words. A word does not become actually ambiguous until it is used in some particular assertion, and so the soundest method of avoiding ambiguity must, like the Socratic request for a definition, keep always some particular assertion in view. It is in this way that we reached ' the doctrine which this book is mainly intended to suggest to the reader's approval, that the validity of all dis- tinctions is relative to the purpose for which they are used at the time — a truth that cuts two ways. It is aimed, on the one hand, at repressing our excessive belief in those distinctions which are (if only for the moment) invalid, and, on the other hand, at enabling us to justify, for a passing purpose, dis- tinctions which are faulty on the whole. There is no distinction which is quite safe against being broken down ; " but there are often strong reasons for hiding this fact from ourselves and our neigh- ' Chap, xii., p, 143. GENERAL SUMMARY 235 bours in given cases, at least for the moment. Hence distinctions are exceedingly apt to seem valid when not so, and unnecessary chains are thus forged for thought. We have to learn that the validity of distinctions is a matter dependent on changing purposes of our own ; that we make, voluntarily or involuntarily, the sharpness of all distinctions that are sharply drawn ; and that we cannot defend it against the'sceptical attack except upon grounds of convenience. The general con- sequences of this view may best be seen by review- ing the chief incidental aims of this book — re- flections on controversy, on the faults of language, and on the conflict between the rival ideals, faith and doubt. As said in Chapter I., some express reference to controversy was almost unavoidable. Indeed, the chief practical results of recognising the rela- tivity of all distinction to purpose are shown in controversy, just as controversy is the chief field of examples of ambiguity. If we once take to heart the lessons derived from our extended view of the faultiness of distinctions generally, and especially of those that most divide us into hostile camps, we shall never again get back, or never quite in its original form, our natural childlike faith in the foolishness of our opponents and the faultless 236 DISTINCTION wisdom of our own beliefs. So hard a lesson cannot indeed be learnt all at once ; but our hope is that by degrees, as the artificiality of all distinc- tions becomes a more familiar fact, we shall come to suspect more and more the opposition between ideals, and, in place of the older and simpler desire for a one-sided victory, we shall substitute the desire to harmonise the dispute by seeking how to limit each ideal by its opposite. For that is always the true centre of philosophical interest in regard to rival ideals — not how to make one devour the other, but how to find room for both ; not how to carry this or that abstract quality or principle to victory or defeat, but how to combine it most judiciously with other and conflicting principles. In proportion as the notion of any virtue becomes definite and determined it stands out by negation of other qualities which also deserve an honourable place in our esteem. To ask which of such oppo- sites — e.g. impassioned zeal or ' sweetness and light ' — is of higher value, is like asking which blade of a pair of scissors does most of the cutting : a given pair of scissors may require one blade sharpened or renewed, so that for the moment that is the one to which attention should be directed ; but a one-bladed pair of scissors would be no more absurd than a man who was all culture GENERAL SUMMARY lyr or all conservatism, or a virtuous person in whom any one virtue existed in such excess as to starve the others. To be well-balanced implies a certain departure from each of the separate ideals. As already noticed, one great remedy for these ' notional ' disputes is the demand for definitions. The purpose of definition, as we have remarked so often, is to obtain a means of applying the dis- tinction in actual cases ; and in the search for actual cases, i.e. for concrete illustrations of abstract qualities, the parties to the controversy must be dull indeed if they do not soon begin to recognise that things in Nature are never abstract, and that if they were so they would be too bloodless and shadowy to excite our admiration and interest. Whenever we find an ideal held up for reverence, the true centre of interest is as to the conditions (the here and now) of the need of. paying exclusive attention to it. Never is the need more than partial or passing. The value of the line between any ideal, and all others that conflict with it, exists — like all other distinctions — only for a passing and limited purpose. And along with our recognition that our oppo- nents', view is not, after all, so different from our own, there will come a great change in our contro- versial method. Instead of taking our opponents 238 DISTINCTION at their worst, we shall find it more instructive to take them at their best ; if we still think it worth while to ask for definitions, we shall no longer care to make use of the insinuations of the Socratic method, and if others attempt that method with us we may simply disarm it by using ' reference- names' — that is to say, by confessing our inability to define, and asking if there is any sense in which they will be content (if only for the sake of argu- ment) to let us use the word in question. Genuine concession, the opposite of assumption, is the root of this method. But to accomplish the best of its work it must be genuine ; its object must be not to defeat the ' opponents,' but to use their minds as well as our own for the discovery of truth. No imitations of it, no self-seeking with sham conces- sion, will serve this better purpose. The progress of the change in us will be marked by our gradual loss of interest in the personalities of dispute, and especially by disuse of the practice of ' scolding,' a weapon which is already becoming obsolete. The chief assumption that underlies this mode of controversy is that some truth and some error is to be found on both sides of every ideal dispute, and hence that the interesting problem is not to find which side is ' right ' and which side ' wrong,' and not even to find which side possesses most of GENERAL SUMMARY 239 the truth — but to sift the truth from the error in both the opposite views. As admitted at the beginning of Chapter IV., there are limits to our power of doing this — namely those set by our natural faculties and training, and by the depend- ence, so far as it goes, of our mental upon our physical states ; that is to say, there will always be a range of subjects upon which we shall have to agree with our neighbours to differ from them. As between old and young, for example, it is notorious that each party has its special immunities and limitations, and similarly as between the active and reflective temperament, or the man and the woman, or the soldier and the tradesman, and so on. But such ' agreement to differ ' is proverbially harmless and satisfactory compared with the differ- ences where no such agreement is reached. A dispute reduced to these dimensions is practically laid at rest. And as to the sifting process itself, some useful machinery for this is provided by the notions we have employed in discussing the subject of distinc- tion generally. When we recognise that criticism of judgment is criticism of distinction we very soon learn to look out for the distinctions that underlie the judgments whose hidden truth and error we are trying to disentangle ; and as soon as the distinc- 240 DISTINCTION tion is found, the question arises in what way the actual interpretation which the judgment puts upon it falls short of the idea which the distinction tries to express. If Nature be continuous, the actual will always fall short of the ideal somehow, and the interesting question always is, not how far short it falls, but on what sort of occasions the shortcoming is important and whether the present occasion is one of them. If these questions are fairly and patiently dealt with, the disentanglement of the truth from the error will be — not finally accomplished, so long as we remain fallible, but advanced as far as the present state of our knowledge permits. A useful guiding idea in our dealings with any disputed question is the attempt to see how far it is ' only a question of names.' And our power of seeing this is greatly increased by our readiness to take all names, not as we should like to apply them, but as our neighbours in fact intend them to be applied. However a name be defined there is always some assumption involved in it, and this assumption may be criticised if we choose — anybody can throw doubts upon an assumption —but it may also be left uncriticised. If we choose the former plan we may keep controversy alive for ever, without getting down to the facts which are GENERAL SUMMARY 341 in dispute ; a better way of reducing a question to fact is to let false assumptions show their own falsity, by adopting them and seeing where they lead. As a rule, indeed, we shall find that our opponents' views do not lead into pure absurdity. As expressed and put forward by the opponents, and especially by opponents who are full of the spirit of partisanship, they will probably have much absurdity in them. But the more interesting matter generally is not the problem how best to bring our opponents' follies to light, nor even how to find what party- name properly belongs to them, but rather to understand as well as we can how their views were reached. It is not, for instance, the bare fact that a man is a Conservative, or a Radical, that is of chief importance (except at the moment of giving his vote), but the process by which he became and remains one — the sort of Conservative or Radical he really is. The second of our incidental aims was to obtain some general views about the nature of language ; to understand difficulties of expression and the way they are met and avoided, and to understand the way in which language acts as a drag on the progress of knowledge through our natural slavery to words. It is specially the clumsiness of words which our reflections upon distinction have brought R 242 DISTINCTION to our notice — the difficulties of expression that arise from the fact that things spoken of are always more full of change and movement than the words we can use in speaking of them, and the consequent tendency of words to hide from us the real com- plexity of Nature and to keep us from seeing it clearly even when, through the progress of know- ledge, we have begun dimly to suspect it in a given case. It is one thing to recognise these difficulties and dangers, but quite another thing to find a final remedy for them. The slack expressions that remain in a language most persistently are those that are most wanted there in spite of their faults. The same permanent conflict that we have seen to exist between the common-sense and the philoso- phical methods in general, exists between the rough meaning of words which suits a dull or hurried or careless insight into facts and the finer meanings that come as our knowledge grows wider and richer in details. Still, a great step is made as soon as we re- cognise clearly the nature of this conflict and the reasons for its permanence. When once we have learnt to regard words as instruments of expression — not given from heaven ready made, but as much our own invention as any tools are — they will be- come our servants, and not be our masters any longer. Thereafter language in general seems to us GENERAL SUMMARY 243 more open to criticism, more improvable, than when we supposed that ambiguous sayings deserve the same sort of respect as the Delphic oracles used to receive. If an assertion is not understood, we are beginning now to see that what it requires is alteration, not ' acceptance ' in its unaltered and useless form ; that what assertions say is never so important as what they mean, and that to interpret a meaning is always to put it into other and; less ambiguous words. The habit of remembering that any case of A is only ' so-called,' and that its right to the name A is always disputable, alters the aspect of the world for us greatly, and helps us to rest our faith rather on facts than on words. Nature, instead of being a museum of specimens whose names are to be simply learnt, is seen as an endless collection of difficult and interesting problems, the answers given to which at any period, though better perhaps than at any previous period, still contain much error which remains to be corrected by degrees. Every name and every distinction is seen to be a nucleus of problems, and even the older and less artificial ones are seen to be full of problems always partly, never completely, solved. This vision helps us also to free ourselves from the wrong kind of fear of self-contradiction — that is to say, from allowing 244 DISTINCTION the barriers made by language to check or stifle enquiry into the way things really happen. We learn by degrees that language is mainly a store- house of old and imperfect theory, a record of early attempts to deal with to-day's problems from a lower general point of view than is open to us to- day. Take the view of the world which language presupposes, and at any date it is somewhat of an artificial simplification of the facts of the world as they may at that date be seen by a mind that is not under slavery to words. The isolation of every * thing ' is fictitious ; in Nature all things depend upon one another, and melt into one another, so that we cannot in perfect strictness say where one 'thing ends and another begins. If knowledge is progressive, this means that at any given date there is something faulty in existing knowledge as crystallised in language. Gradually we shall come to regard unreal distinctness as a source of error that is always to be taken into account whenever we use a name or distinction ; the only question in any case being to what extent the objection is re- levant here and now. That words are essentially instruments of ex- pression is a formula that will be found useful. It is intended chiefly to contradict the view that words are essentially names of things — of some- GENERAL SUMMARY 245 thing real — a view which leads us to exaggerate the fixity of their meaning and the extent of its acceptance in one clear sense by our neighbours. As we have seen, ' unrealities ' — things which we all admit to be unreal — are named as freely as we name anything at all ; and, while a named reality ' may rightly suggest the question, ' What is the thing so named ? ' a named unreality often leads us very far astray if it suggests that question. The actual things that most nearly correspond to abstract notions are always easy to caricature ; they never quite exemplify the ideal. And therefore, unless the names are taken lightly,' by tact of one sort or the other, the thought they attempt to express is apt to be unfairly treated. When we demand producible examples of ' culture,' ' truth,' and so on, we are on very firm ground, controversially speaking ; some fault can probably be found with any example produced ; but whatever indirect value the sceptical question may have, as drawing attention closer to the facts, it never has so much direct importance as the mere controversialist imagines. The last word as to a prophet's teaching is not said when we have found his doctrine somewhat hazy in its details. And another familiar view of the nature of ' I here merely concede that there is any such thing. 246 DISTINCTION words is contradicted by regarding them as instru- ments of expression — the view that they are essen- tially ' signs of ideas '; or, rather, that there is always someone idea which corresponds to a word, our business being to find it. If words are instru- ments of expression, our business is rather to find the word for an idea than the idea for a word ; words mean only what those who use them choose to make them mean ; and instead of trying to tie a speaker down to the usual, the oldest, or (what we think) the best, sense of a word, the interests of truth are better served by getting him to tie us down to the meaning he intends. Then and only then will the assertion that we criticise be really his assertion. On its negative side the truth that words are instruments helps us to become aware of the extent of unreal distinctness which exists at all ' times in ordinary thought. But its effect is not merely negative and critical. Its positive side is seen especially in one of its consequences — in the im- portance it gives to the practical purpose of words — to the search for the reason which the given speaker may have, just then, for choosing that particular word in preference to any other. Thus it familiarises us with the notion that all the faulti- ness of language is merely an outcome of a conflict GENERAL . S UMMAR V 247 of needs, and that if we can manage to keep words sufficiently under control, sufficiently free and elastic, we may use them to meet the need of the moment with some success. ' We learn, that the only valid excuse for unreal sharpness of dis- tinction is in connection with special and passing purposes ; but also that this is capable, on occasion, of becoming more than a mere excuse — that on occasion the unreal sharpness of a distinction may be advantageous — like the work of an artist in painting — the selection of the most important points which shall make a picture vivid, or make fiction truer than fact. On the whole, the two notions that I think may help us most in reaching this view of the nature of language are that of the constant tendency of words to get spoilt (Chapter VIII.), and that of the connection between abstractness and vagueness {Chapter X.)., As regards the former, it takes most of us half a lifetime, or longer, to learn to distrust words enough on account of their associations. Epithets in common use are the typical kind of words that get spoilt ; and the natural easy-going opinion is that such words, just because everybody uses them and ' everybody knows what they mean,' are excellent verbal currency. The reverse is truer. Just because they are applied by many 248 DISTINCTION minds possessing genius in different degrees, the highest genius being ever the rarest, and because n the great majority of cases their application is made in a hurry and without much care, the worse applications outweigh the better ones increasingly. The A's which are A on the surface are those which common-sense inevitably learns to associate with the name, not the A's which are A in the best sense of the term ; so that, for instance, the man who throws money about is the ' generous ' man, and he who can trample is ' strong.' It is. hardly too much to say that the commoner an adjective (or an adjectival substantive or verb),, the less can it be trusted not to mislead our thoughts. And as regards abstractness, the earliest notion we get of this is, probably, that the abstract is higher than the concrete, as being more refined or more difficult to understand. Whether there be any sense in which this, or something like it, is true we need not here enquire, since at any rate the opposite view is equally true and instructive. The thinner and scantier our knowledge of any subject is, the more abstract are the statements we make on that subject. All progress of science includes a progress from more to less abstract assertion — that is to say, to a fuller specification of GENERAL SUMMARY 249 the conditions of inference from facts as named. As knowledge grows we find this and that general assertion a little too broad, too vague, or loose ; it is true ' in the abstract,' but for' practical purposes we need something more than merely abstract truth. ' A,' for instance, whatever it stand for, is never found quite pure, but always entangled with other qualities (B or C) which either hinder or help the inference from A in the abstract. And in proportion as we discover the help or the hind- rance that B or C render to A in this respect, the name A, unqualified, becomes clumsy for practical purposes. But if knowledge is throughout pro- gressive, then every name that we use for stating conditions of inference will admit of being refined by qualifications, and the interesting problem always is to find in special cases what exactly are the qualifications required. The more we become impressed with this need, the less we shall trust words uncriticised, and the more we shall seek to correct their shortcomings by getting closer to the facts. Description is always partial description, and the individual case is always richer in detail than is shown by any name or string of names we can give to it. Lastly, in regard to scepticism, it is not easy to condense to any purpose what was said in Chapter 2SO DISTINCTION XVII. Rather, the subject needs much expansion before the two spirits that war within us will come to anything like a final understanding. It may well be doubted whether any preachings can help us greatly towards attaining in our character the best proportion of positive and negative elements, or the best combination of common-sense and philosophy. The lifetime of everyone is a succes- sion of private and mostly unconscious attempts to suit that part of his character which lies within his power to alter, to that part which is beyond it. General statements on the subject, such as pleadings for or against scepticism, are likely to seem stale and trivial to anyone whose mental growth has been troubled by thought. Those who understand the past at all can hardly help condemning much of the modern enthusiasm for destruction ; those whose sympathies are modern can hardly find refuge in the old methods of ruling out unbelief. Both parties admit that doubt is in the air to-day more than at any past time ; and though practical life, with its need. for happy assurance, remains for us all the most pressing and permanent of realities, our wish to preserve that happy assurance is dis- tracted by an equally tempting wish to preserve our honesty of mind. The followers of these two distinct ideals may easily scold each other for ever, GENERAL SUMMARY 251 but for those who try to harmonise them there is perhaps a brighter promise of results. The plan of enrolling ourselves under rival banners has surely been tried long enough. Can we not now leave our opponents to find what name they please for us — sceptics or otherwise — and so get free to attack the impersonal question in a rather less prejudiced way ? No one else knows our inmost beliefs so well as we know them ourselves, and yet we ourselves do not know all their firmness, and all their vacillation, perfectly. And at any rate the question ' What is true .' ' is of wider and more lasting interest than the question what beliefs do you or I at present claim as our own. APPENDIX Nameable Things (see p. 14). That every name, as such, must be the name of some- thing seems at first sight to be a too obvious truth. To name something or other — one thing or many — is natu- rally regarded by common-sense as the most essential function of a name. This common-sense view, as usual, rests upon a truth, but obscures it through careless expression, and in order to substitute for it the truer view that names are essen- tially instruments of expression, it seems better not to deny that they are also essentially names of ' things,' but to render the admission harmless by adding that, in any sense in which the distinction real-unreal can be used, the ' things ' that are named need not be real. They need only have reality in the sense in which ' unrealities ' have it — that is to say, only when the word reality is entirely deprived of its meaning for want of a background. This harmless admission seems best made by means of the phrase ' nameable things.' Names, as such, are the names of nameable things, but we may name unrealities when- ever we please to do so. The ' thinghood ' of nameable 254 DISTINCTION things is a convenient fiction, and its convenience need not prevent our seeing it as a fiction. The meaning of a name is not something distinct from its use, though its common meaning or use may of course be distinguished from its use or meaning on a special occasion. Whatever meaning names have is only as entering into assertions. And hence the very type of a meaningless name would be a name which could never be used as a predicate, could never be applied to anything at all. In a given case, no doubt, we may find it difficult or impossible to decide on the proper application of a name, but the name is supposed to be predicable never- theless, however difficult to predicate correctly in given cases. If we recognised no such possible thing as truth, the purpose of the name would vanish, just as much as if we recognised no distinction between truth and falsity. It may be asked, 'Do names like centaur, gkostj and miracle become meaningless as soon as we arrive at the belief that there are no such things ? ' The question is useful in forcing us to beware of a slight ambiguity in what was said just above. For brevity, it is perhaps legitimate enough to say that the meaning of a name dis- appears as soon as we admit the non-existence of the thing it denotes ; but the statement certainly is not true in the sense which the question assumes. Obviouslyithe very proposition which denies the existence of the thing in question uses the name as if it had a meaning ; if the name miracle had no meaning, the assertion that miracles do not happen would lose its meaning also. More strictly, then, what is necessary to the meaning of a name is not a belief in the ' real existence ' of the thing denoted, but the bare concession, on the part of those who are using APPENDIX 255 the name, that there are (real or imaginary) things or cases to which the name may be applied ; the name must be supposed to be predicable of something or other. It is not necessary that we should allow in a single actual case that the name is correctly predicable, only that we should regard it for the moment as possible that such cases may be found. The habit of regarding names as essentially names of things both encourages and draws support from an over- simple theory as to the origin of language. The earliest origin of names cannot rightly be conceived as vocabulary- making— that is to say, as simply prior to assertion and independent of it. Strictly taken, the question as to the earliest origin of names lies beyond the re^ch of any direct historical or even philological evidence. We can, there- fore, only do our best to imagine, however negatively, the most conceivable beginnings of the process ; and then it soon becomes evident that in some way or other we must manage to avoid that easy view of the relation of names to assertions which regards the latter as built out of the former as a house is built of ready-made bricks, or as existing names may now be put together to form an as- sertion. It is safe to guess, at least, that primitive man did not form his vocabulary as a finished product before he began to assert, did not name things for the mere sake of naming them or even of using the names at some in- definite future time, but that the process of asserting and that of naming must have gone hand in hand from the outset, names arising only in order that this or that asser- tion might be made, and so being primarily instruments of assertion. We may safely suppose that every system of names has been evolved in much the same way as any 2S6 DISTINCTION organic instrument — -e.g. the nervous system — in and by its operation, and not by far-seeing attention on the part of primitive man to the possible needs of his remote de- scendants. Thought and Language (see p. 14). Distinction, it was said, implies a recognition 'in thought and language ' of a difference between one name- able thing and another or others, and it may be asked whether language and thought are supposed to be co-ex- tensive. That entirely depends on how the terms are defined, and we may simplify the problems of distinction greatly by waiving the question whether there is anything that can be called ' thought ' in the absence of language, and by considering in the first place only the part that is played by distinction in language and in so much of thought as is expressible in language. To prevent need- less misunderstandings, however, there is no harm in confessing that on this subject I side distinctly with those who see no reason (beyond regard for the roughest practical purposes) to limit the term ' thought ' to such thoughts as correspond to ready-made concepts. We may of course so limit the meaning — we may define our terms as we please — but the desire to define thought as finished thought, or highly-developed thought, seems to point to a survival of the unhistorical spirit, even in those who, like Professor Max Miiller ('Science of Thought,' and ' Natural Religion '), openly claim to belong to the ' his- torical school.' For as soon as we ask how new concepts actually arise, the shortcomings of this artificially narrow definition become evident. A new concept exists, how- ever indefinitely, before a new name begins to help it to APPENDIX 257 struggle into definiteness ; the first glimpse of the need for a new name is clearly prior to even the first partial satisfaction of that need. Only if no new concepts are ever gradually reached by man, can thought begin simul- taneously with language. TTiinghood {ste. pp. 14, 67). It is hard to imagine how the substantive would ever have been distinguished from the verb and the adjective unless we had had the conception of passing states in that which is relatively permanent. Relative permanence is, however, suificient, without raising any awkward questions as to the deeper meaning of ' thinghood.' In order to be viewed as substantival, it is not necessary that a ' thing ' should be everlasting and immutable ; it need only, as a whole, be capable of outlasting some partial change. It is because the tide may be high or low that tide is a.sub- stantive, and because the height of the tide (or of any- thing else) may vary, that height is a substantive too. The substantive is essentially the name of that which may be regarded as having attributes {i.e. as varying in this or that respect — e.g. high and low tides, more and less height), and the question whether itself is a thing or an attribute is irrelevant-the fact, either way, even supposing it cer- tain, is accidental. And so abstract names, in the sense of substantival names of attributes, always arise when we regard any action or quality of things as itself having attributes, as capable of any quantitative or qualitative variations whatever. The notion of change carries along with it the con- trasted notion of permanence, or stability. By a ' change ' we always mean a change of or in something that stands 258 DISTINCTION relatively still while the change goes on. No change can ever be recognised as such except by the aid of a back- ground that changes less, or does not change ; no change can appear to take place unless there appear to be some- thingthai changes — something that, after the change, exists in an altered form. This something may be as imma- terial as you please, yet its name is essentially substantival. To distinguish between ' things ' of this sort and things in any more satisfying sense demands (even if it be pos- sible) a more metaphysical enquiry than people in general are wont to give to it. And accordingly, since every event is a change or a set of changes, while events are the sum and substance of our experience of Nature, our notion of Nature as a whole grows to be that of a concourse of things which are liable to alteration. We conceive as ' things ' the rela'ively permanent or stable element in Nature ; and contrast them with the ' attributes ' of things, which are relatively unstable. We soon learn to admit that the permanence of things is only a convenient fiction, that all the things we know are destructible — that is, are capable of suffering so much change that their identity disappears. Things, as we know them, become and de- velop and melt into something else. A cloud or a puff of smoke is a thing, yet it spreads away in a minute ; a mountain-range is a thing, yet the weather will slowly wear it down. And even before a thing disappears or becomes finally decomposed, we are aware that our habit of calling it. the same is somewhat of an artifice ; the river that remains is not the water that passes by nor the bank that crumbles, and in our friends and ourselves that in- most personality which is the same to-day as in childhood is confessedly immaterial. APPENDIX 259 To use any name substantivally, then — to form a substantive — is to regard as a 'thing' that which the name denotes, whether or no we may at other times regard it as an attribute of something else. There is thus a sense in which it is impossible to avoid ' realising our abstractions,' or giving them a fictitious existence. But we may regard this or that as a thing without believing it to be so ' really ' ; we need hold no metaphysical theory as to the nature of 'thinghood.' All that is necessary is that between the parties using the language there should be a tacit agreement that . for the purpose in hand it is irrelevant to notice the fleeting or negative or dependent character of the ' thing ' so named. Grant that a headache is nothing except a passing state of a sensitive subject, still we may easily agree to call it real ; grant that a defect, or a minus quantity, is relative to an imaginary something from which it is to be subtracted, still it is something that may be spoken about with plenty of practical meaning. Shifting and negative as they are, to speak of them at all is to give them an artificial steadiness or independence, as the photograph fixes the lightning or the momentary look of a face. Only by some such artifice can our thoughts attempt to deal with the fleeting facts of Nature. The Continuity of Nature (see p. 72). The deeper difficulties of this subject were purposely avoided in Chapter VI. They consist chiefly in making clear to ourselves the notidn of continuity. It may be asked, for instance, Does not the argument that we cannot conceive of Nature as discontinuous lose some of its S2 26o DISTINCTION force when we try to conceive evolution itself? Such a conception is also just as impossible as that of creation ex nihilo. Evolution is at any rate change, however gradual ; and, subdivide the time as much as we will, we cannot conceive of any period of continuous change so short that no change takes place in it. But change, as such, is a saltus, or a creation ex nihilo, and is none the less miraculous (or inconceivable) because it is small — for size is a purely relative notion, and depends upon the point of view. Is it not just as true, therefore, to say that Nature is eternally discontinuous as to maintain the opposite ? Instead of there being no break or seam in Nature, is not its whole structure composed of breaks or seams — which only do not appear such because they are small and familiar ? This objection may help us, at any rate, to understand the dependence of meaning upon agreement. If the question, ' Is Nature continuous throughout ? ' is to have any meaning, we must agree to give it one somehow ; that is to say, the notions of continuity and discontinuity must be somehow contrasted. It may be true, as the o"bjection implies, that the contrast between these notions is open to criticism. Only, if we decide to criticise it, instead of adopting it 'for the sake of argument,' then we can no longer ask the question intelligibly, since neither answer — yes or no — has any preference over the other. So regarded, the solution of the puzzle consists in translating the question, ' Is Nature continuous or not ? ' into other and less ambiguous terms. Its purpose, in the text,' is to ask whether between so-called A and so- ' P. 72. APPENDIX 261 called non-A thfere is, or is not, always an intermediate region ; and to this I answer that we cannot possibly conceive the absence of it. Moreover, we have much experience of its unexpected presence, since no experience is commoner than to discover that some so-called A or non-A only roughly deserves to be so called. Look closer, and your so-called ' men ' may be seen to be descended from ' beasts ' ; look closer, and your so-called ' straight ' line may be seen to be a little crooked. Besides, what meaning can be put into the objection imagined above, except that it raises the question whether between A (or non-A) and the intermediate region there is not a perfectly shacp distinction ? And such a question cannot on our own principles be answered in the affirma- tive. Ideally, no doubt, the distinction is perfectly sharp — but so is that between A and non-A themselves ; on what ground are we to suppose that this new ideal contrast is any better actually ? The difficulty of separating the intermediate region from either extreme is precisely the same as that of separating the two extremes from each other. A small change may be no less miraculous than a large one, but the same is surely true of a small gap. The gap between A (or non-A) and the intermediate region, small though it be, demands to be filled somehow. Mill's ^Inductive Canons' (see p. 128). These are sometimes supposed to be general rules of fiict-evidence, under which the special case may be brought. So regarded, they pretend to say that if under certain conditions we notice certain points of agreement and difference in ' two or more instances ' observed, we 262 DISTINCTION can then reach a conclusion which, if not absolutely certain, is at any rate practically certain, and worth calling a ' valid induction,' as contrasted with a ' merely empiri- cal law.' As general rules of fact-evidence, their weak point lies in the description of the conditions, which are worded so as to be either inapplicable to actual cases or else worthless when applied. If we interpret the Canons in one way, we cannot be sure that the precautions have in fact been properly taken, in a given case, until after we have otherwise arrived at a full knowledge of the causes concerned ; if we interpret them in any other way, they may lead us right in many cases by accident, but in many other cases they will, be sure to lead us wrong. Look, for example, at the Canon to which Mill himself refers, more than once, as possessing 'rigorous cogency' above all the others, and as enabling us to 'arrive with certainty at causes' — the Canon of the ' Method of Difference.' If an instance in which the phenomenon under investigation occurs, and an instance in which it .does not occur, have every circumstance in common save one, that one occurring only in the former ; the circumstance in which alone the two instances differ is the effect, or the cause, or an indispensable part of the cause of the phenomenon. Thus, if we make or notice any single alteration, followed by a further single point of difference, in an existing state of things, the state of things before and after such alteration provides our ' two instances ' ; and the further difference is the effect of our single alteration. For example, if we are investigating the cause of the stoppage of a watch, and we find something stuck in the hair-spring, on the removal of which obstacle the watch at once goes on, the stopped watch is the APPENDIX 263 ' instance in which the phenomenon under investigation occurs,' the going watch is the ' instance in which it does not occur, and to the best of our knowledge the former and the latter instances ' have every circumstance in common save one ' (the obstacle in the spring), ' that one occurring only in the former ' ; hence, the said obstacle is ' the cause or an indispensable part of the cause ' of the stoppage. Strictly taken, the condition under which this Canon can be applied to a special case is one that may be fulfilled without our knowledge, or that we may think fulfilled without its really being so ; but which we can never securely know to be fulfilled in a given case until we already know the causes concerned. What leads us to open the case and look at the hair-spring? Our knowledge, however scanty, of the works. Suppose a clock stops at the moment of someone's death ; a less instructed person might there get two instances appearing to him as convincing as those just mentioned above. To a savage, a European who orders an eclipse of the sun at the moment when the almanac tells him to expect it, is clearly ' an indispensable part of the cause ' of that phenomenon. And no one knows better than scientific men how easy it is to misread our most careful experi- ments. Watch a conjurer at work, and try to apply Mill's Canon ; you will find that his whole art consists in leading you to believe that you take suflScient pre- cautions when you ' observe ' that only one circumstance has varied, while he himself stealthily varies another behind your back. The difference between Nature and a conjurer in this respect is only that Nature seems not to care whether we are deceived or not. 264 DISTINCTION Some Examples from Philosophy (see pp. i6o, 217). Those who are just beginning to study the history of philosophy will find it a useful plan to acquire the habit of regarding all philosophical questions as con- cerned with the application of ideal distinctions to actual cases. The following notes will perhaps serve to give the student a start in this direction. Since Descartes, philosophy has been much con- cerned with the distinction between self and the ' outer ' world— has attempted to criticise the sharpness with which that distinction was formerly conceived. We can hardly get far along the path of ' observing ourselves ' without raising the question, How is it possible for the thinking subject to be its own object ? Ideally, subject and object are in perfect antithesis, but how about the actual contrast ? Locke assumes, quite simply, that ' we ' can ' study our own ideas ' ; and, though Berkeley began to find the distinction between esse and percipi unsatisfactory, it was necessary for later philosophy to take some trouble in pointing out that the ego is not merely an object (non-A), but that so far as known (or observed) at all it is known by the ego acting as subject (A). The pure subject could, from the nature of the case, never be known, since in being known it becomes object. Descartes' criterion of truth — clearness and distinct- ness—is open to the charge of being an ideal one, not actually applicable. Any proposition, he says, is true if the necessity of connection between its parts is as clearly and distinctly perceived as in cogito, ergo sum. But what actual propositions are in this condition ? Ideally, ' it is APPENDIX 265 impossible for God to deceive us,' but actually we are often somehow mistaken. Ideally, this may be, as he says, the fault of our will, not of our reason, but actually we cannot be sure in given cases whether it is reason or will that leads us. And in general this was the constantly-recurring fault in Descartes' meta- physics. He was apt to be content with truisms — i.e. did not trouble sufficiently about the difficulty of applying truths in actual cases. Thus, in the cogito, ergo sum, the undeniableness depends on the vagueness with which ^ sum' is conceived.* Descartes' habit of arguing from definitions to facts is another outcome of his failure to distinguish the ideal from the actual ; ' the soul thinks always ' (as a fact) because the word (or idea) ego includes the idea of consciousness ; or again, God exists (as a fact) because in the idea of God existence is clearly contained. And his four rules of method are almost worthless as practical guides — worthless, except possibly to children ; that is to say, they are truisms,^ and in order to put a practical meaning into them we have to suppose them said to someone who needs to be told (i) that error is possible ; (2 and 3) that analysis, up to a certain point, is useful, and that the simple is more intelligible than the complex ; (4) that we may easily ' Malebranche was just as certain that we exist, but found more difficulty than Descartes in regard to the question what our exist- ence is. - (Paraphrased) : — (i) To avoid undue haste in judging, (2 and 3) To proceed as gradually as is necessary, (4) To make enumerations so complete, and reviews so general, as to satisfy myself. Does not every mistaken person fancy he applies these rules ? 266 DISTINCTION overlook some relevant detail. Such rules are doubtless ideals, but as practical guides they tell us nothing at all. Again, Locke's criticism of the use the Cartesians made of the notion of innate ideas amounts to saying that the distinction there drawn between primaiy know- ledge and knowledge derived from experience is inapplic- able to the actual knowledge possessed by the human mind. If we look at any actual knowledge, he says, we find it is dependent upon ' experience.' And the later critics of Locke have in effect brought a similar charge against Locke himself: the inexperienced mind in the abstract may certainly be, as Locke says, a tabula rasa, but so abstract a truth does not help us, since no actual mind is found in this condition ; and, indeed, if it were so, experience would be impossible to it. Before any experience (A) can complete itself, we must pass through stages at which it is incomplete (non-A), and nowhere can we either find or imagine its absolute beginning. So, again, the celebrated answer of Leibnitz ' to the assertion (adopted by Hobbes and Locke) that there is nothing in the intellect which was not previously in the senses, implies the same kind of objection in another form. A philosophy (it says in effect) which sets intellect and sense in sharp opposition to each other, fails to raise and answer the doubt whether after all the distinction between them is not somewhat unreal. For if all actual sense-perception (non-A) has some intellectual element (A) in it, the derivation of ' A ' from ' non-A ' loses its ' * Nisi intellectus ipse. ' APPENDIX 367 power as an explanation of the general origin of A, and becomes only an explanation of the origin of what we call ' A ' from that which we wrongly call ' non-A.' And any assertion that A has arisen out of ' some- thing else' lies open to the same criticism. The fact that ' A ' has become cognisable as A is proof that the germ of A {i.e. A in a less cognisable form) was there before. But between A and the germ of A— between A as more and as less cognisable — no line is possible but an artificial one. The fact that any so-called non-A contains the germ of A is proof that the name non-A is there in strictness wrongly applied- Middle Term and Major Premiss (see pp. 190, 231). These technicalities, like many others that are used in logic, are explained very superficially in the books whose purpose is to prepare students for examination. For that purpose it is generally thought best to define all such terms by reference to the scholastic logic in its most unyielding form. Thus the student is told to find the middle term of a given syllogism by noticing which term does not appear in the conclusion : and the 'major premiss ' is ' the premiss which contains the major term ' — the major term being ' the predicate of the conclusion.' So that if our conclusion happens to be written {e.g^ ' Some X are Z ' instead of in the equivalent form ' Some Z are X,' the major premiss is the premiss in which the term Z occurs as predicate, whether it contains the statement of a general rule or not. In a large percentage of cases this mechanical way of 2-68- DISTINCTION finding the middle term and the major premiss may suffice.' Although it can never throw light on the process of syllogism, it will often not have any direct misleading influence. But in proportion as we recognise that assertions get their character, not from the accidents of their form of expression, but from the purpose they are intended to serve in some particular argument, we are driven to find some other way of distinguishing major and minor premisses. In a treatise directly on logic it would be necessary to discuss this question at some length, but here perhaps the following hints will suffice. Think of any assertion, put before us as requiring proof. Such, proof always consists in bringing forward facts as signs of its truth. But what makes such facts sufficient for proof? Always our knowledge, so far as it goes, of the regular ways of Nature. A fact proves a conclusion to us just so far as we take it as a trustworthy sign of something — smoke, for instance, as a sign of fire. And it is precisely these sign-beliefs ('inferential' or 'conditional' or 'general' judgments) that perform, when so used, the function of major premiss ; the state- ment of the fact is the ' minor premiss ' ; and the ' middle term ' is the term through which the major and the minor premiss are connected. If we ever prove fire by means of smoke, the fact ' there is smoke 'is the minor premiss ; the rule of inference ' where there is smoke, there is fire ' is the major premiss ; and the middle term is ' smoke.' This view of the nature of syllogisms is not really an ' As a matter of fact, three out of the nineteen ' valid moods of the syllogism' are stultified by i\.:—Disa»iis, Bokardo, and Dimaris. APPENDIX 269 innovation, though it is apt to be too much hidden by the verbal trivialities of the older logic. There will never be life in the study of logic till we learn that words are only counters ; and probably the scientific men are those who are now most effectually teaching us this lesson. INDEX ( Words in Italics refer to Examples. ) ABS Abstract names, 49, 62, 69, 85-6, 17s, 229, 24s, 257-9 — assertion, 21, 99, 121, 129, 248, 249 Actual and ideal, 27, 34-5, 47, 49. S3. 56. 66, 8s. 117. 121. 129, 170, 185 »., 222, 240, 261, 262, 264 7 Adjective, substantive, and verb, 5. 67, 173-S. 257 Admitted reality, 58-70, 85, 112, 131, 141, 187, 213, 227, 24s Agreement and meaning, 112, 181, 216, 223, 232, 240, 259; 260 ' — provisional, 60, 141, 187, 211, 218 Alternatives and distinction, 144, 164, 198-200, 217 Ambiguity : (a.) its nature : — effective and ineffective, 4, 134, 232 the latter not properly am- biguity, 4, 112 ambiguity (proper) attaches to assertions, not to words, 4. 234 does not depend on extent of borderland, 104- 1 1, 232 AMB Ambiguity : (o.) its nature (cont.) : — most effective where most occasional, 6, 96, 151 rough distinction the chief source, 8, 16, 22-9, 84, 96, 230 clear meaning involves clear distinction, 15,27,99-112, 140, 19s, 198, 231 connection with abstractness, 21, 25, 27, 49, so, 85, 88 «., 97, 99, 121, 129, 237, 247-9 chiefly occurs in speaking of immaterial things, .85, 97, 229 (b. ) its extent : — every general name liable to it. 73. 94. 233, 242 continuity of Nature, 71-9, 82, 84, 104, 198, 259-61 laws of thought, in applica- tion, 56, 71, 109, 117, 217, 230, 243 occurs wherever common- sense (or philosophy) is divided against itself, 82-7, 42-57, 89-97, 228 272 DISTINCTION AMB Ambig«j«y (cont.) — (c.) its effects : — When unseen — 'idealisation and caricature,' 45. 89, 133. 211, 229,236, 245 narrowness of view, 99, 134, 199 slavery to words, 12, 20, 74, 87-97, 183. 217, 221, 241-4, 269 When seen — destruction of meaning, 99-112, 129, 163, 195, 222, 231, 260 (d. ) remedies : — (I.) The Socratic method, 24, 25, 104, 114-30 its strength, 1 15-18, 129, 166 its failure, 1 31-49, 183, 216-24 (2.) Common-sense tact, 9, 17, 30-40, 63, 106 n., 133-47, 225 its value, 38, 82, 94, 135 its failure, 37, 83, 89, 93, 135, 228 (3.) The method of ' special relevance,' 1 1 1, 135, 143, 233 its value, 139, 151, 182-4, 215 its cost, 166, 184-6 Animal and vegetable, 28, 76 Appearance and reality, 215 Applicability of distinctions, 22, 30, 100 »., 121 «., 254, 264-7 Argument — 'for the sake of,' 60, 141, 14s, 187, 211, 218, 260 Argumentative use of a distinc- tion, 144, 164, 217 Arnold, M., 46 Assent and denial distinct, 104, 130, 206, 213, 231 COM Assertion, doubt, and denial, 204, 213, 250 Assertions of tendency, 120-24, 129, 249 Assumption and concession, 141, i8i-9, 2II-2I, 235-9 Assumption and theory, 187-9 Assumptions, criticism of, 35, 188, 191, 213 Attributes and things, 62-7, 85, 97> 141, 257-9 Background of names, 15, 99- 112, 231, 253 Begging the question, 182, 185, 186, 188, 189, 213, 216, 217 Beliefs, criticism of, 10, 113, 150, 193-200 Berkeley, 264 ' Borderland,' extent unimportant, 104-11, 232 Caird, Professor, 215 Cairnes, Professor, 124 Capital, 121 «., 160 Caricature, 45, 89, 133, 211, 229, 245 Casuistry, 12, 22, 25, 106, 114, 185, 204, 223 Causation, 36, 127, 12S, 196, 208, 2&3 Certainty, practical, no Chemical elements, &c. , 74, 182 Child axiA man, 21, 83, 103, 169, 198 n. Clever, 90, 92, 95 Clumsiness of general names, 91, 94, 95. 165. 230, 241, 249 Common sense, as at war with itself, 42-57, 82-7, 89-97, 228 — grades of, 8, 36, 39, 80-7, 93. 133. 225, 248 INDEX 273 COM Common tact, 9, 17, 30-40, 63, 83. 93. 94. 106 «., 13s, 147, 225, 228 Comparison (or value) and mean- ing, IS, loi, 231, 253 Concession and assumption, 141, 181-9, 211-21, 23S-9 Conciliation and scepticism, 202-24 Conditions of inference, 122, 123, 197. 249. 268 Connotation, 100 «., 108, 168, 181, 183, 184 Conscious and unconscious experi- ence, 20, 153 Conservative and Liberal, 45, 98, 241 Conservative tendency oflangu age, 12, 20, 183, 217, 241-4 Continuity of Nature, 71-9, 82, 84, 104, 198, 259-61 Controversy [see also under ' Ambiguity ' (c. ) and (d. )] : — wide meaning of the word, 166, 192, 219 main sources, 42, 51, 55. ^5. 89, ^T, 229, 240 Tricks and methods : — The controversial demand for definition, 12, 24, 25, 51, 5S. 1 13-130. 143. 185. 193-8, 223, 231, 237, 24s The defences against this de- mand, 9, 47 k., 94, 106 «., 132, 13s, 183 ' Idealisation and caricature,' 45. 89. 133. 211,229,236, 24s Accusation of truism, 82, 125, 197, 207, 213, 265 Begging the question, 182, 185, 186, 188, 189, 213, 216, 217 Assumption that a theory is dogmatic, 187 DEN Controversy : — Tricks and methods {cant. ) : — Admissions made 'for the sake of argument,' 60, 141, 145. 187. 211, 218, 260 Waiving a question, 59, 89, 141, 181, 184, 188 Accusation of quibbling, 106, 132, 136, 193, 210 ' Only a question of names,' 28, 52, 91, 240 The concessional method, 2H-2I, 235-9 In general : — questions of fact and of opinion, 41, 155, 194 truth and error always min- gled, 24, 34, 35, 81, 85, 206, 219, 221, 222, 239 doubt stronger than denial, 187, 192, 204, 207, 245 Counteracting circumstances, 121-4 Creation ex nihilo, 72, 260 Criticism of assumptions, 35, 188, 191, 213 — of distinctions and beliefs, 10, 113, 150, 193-200 Culture, 40-50, 55, 67, 85, 245 Cyclone, 61 Darwinism, 77, 157, 161, 188 Degree and kind, 16, 71-9, 161, 227 — of resemblance, 137 Demand for definition, 12, 24, 25, SI. 55. 1 13-130. 143. 185. 193-8, 223, 231, 237, 245 Denial and assent, 104, 130, 206, 213. 231 — and doubt, 187, 192, 204, 207, 250 Denotation, 100 n., 168, 181, 183, 185, 254 T 2/4 DISTINCTION DES Descartes, 264, 265 Difference, essential, 6, 92, 137, 164, 230 Distinction : — Explanation of the term, 14, and of ' rough ' distinction, 1 5, 17, 226; distinction and definition the same process, 15 ; distinctions of kind and degree, 16, 71-9, 162,227; how are rough distinctions possible? 19 ; other ways of characterising rough distinc- tions, 21 ; 'unreal distinct- ness,' 22, 88 (For the nature, extent, and effects of unreal distinctness, see under ' Ambiguity ' (o. ), [b.), and {c.) The ' applicability ' of distinc- tions, 22, 30, 100 »., 121 »., 254, 264-7 Distinction as ' ideal ' and 'actual,' 27, 34, 35,47, 49, 53, 56, 66, 85, 117, 121, 129, 170, 185 n., 222, 240, 261, 262, 264-7 Assent and denial clearly dis- tinct, 104, 130, 206, 213, 231 Relation between criticism of distinctions and of beliefs, lo, 113, 150, 193-200 The relativity of distinctions, 143, 164, 223, 234, 246 Distinction as the creation of alternatives, for use in argu- ment, 144, 164, 198-200, 217 Distinctness, unreal, 22, 88 Dogma and theory, 187 Doubt, force of, 192, 204, 207, 223 Effective ambiguity, 4, 134, 232 Elements, 74, 182 IDE Emphasis on a distinction, 145, 164, 180, 181, 217 ' Empirical ' generalisation, 208, 262 Essential resemblance and differ- ence, 6, 92, 137, 164, 230 Evil, 27, 86 Exception and rule, 36, 99, 109, 122, 127, 133, 135, 147, 226 Excuses for vagueness, 47 k., 89, 94, 106, 117, 121, 132, 143, 147, 183, 201, 223, 233, 238, 242, 247 Extent of borderland unimportant, I04-II, 232 Fact and theory (or opinion), 41, 154. 194, 206 Faculties of the mind, 86, 1 53 General and special relevance, 139, 144. 162, 166, 179, 234 ; and see ' Special Occasion ' — names, 14, 24, 55, 94, 168- 178, 180, 182, 220, 233, 242-8, 253-6 Gerard, Rev. J., 158, 187 Grades of common-sense, 8, 36, 39. 80-7, 93, 133 Grammatical distinctions, 67, 173-5. 179 Green, T. H., 20 ». Horizon, 63 Ideal and actual, 27, 34, 35, 47, 49. S3. 56. 66, 85, 117, 121, 129, 170, 185 «., 222, 240, 261, 262, 264-7 'Idealisation and caricature,' 45, 89. 133. 211, 229, 236, 24s INDEX 275 IDE Ideals, rival, 35, 41-S7, 89, 133. 193. 2". 212. 236, 237, 250 Immaterial things, 60-9, 85, 97, 229, 24s, 253, 257 Independence and 'thinghood,' 65. 2S7-9 IndestrucUbtlity of matter, 72, 258 Inductive canons, 128, 26 1 -3 Inference, conditions of, 122, 123, 197, 249, 268 ' Instruments of expression,' 12, 25. 69, 177, 179. 215, 240, 242, 244-7, 2S3-6, 269 Invention of names, 170, 255 Irrelevance of demand for defini- tion, 131-49, 183, 216-24 Justice, 23«., 25, 117 Kind and degree, 16, 71-9, 161, 227 Knowledge, progress of, 32, 73, 88 «., 94, 123, 182, 221, 241-S, 248, 256 — origin of, 156 Labour and hired labour, 160 Language and its diffi- CtJLTIES : — (a.) The nature of language generally : — Conflicting needs, 11, 95, 242, 246 General name {or notion), 14, 24, SS. 94. 168-78, 180, 182, 220, 233, 253-6 Language and thought, 14, 17, 21, 56, 157, 19s, 221, 241-9, 255, 256 Words as instruments (or counters), 12, 25, 69, 177, 179, 215, 240, 242, 244-7, 253-6, 269 LAN Language and its Difficul- ties (cont. ) : — Names as abbreviated h)po- thetical sentences, 69 Assent and denial must be sharply opposed, 104, 130, 206, 213, 231 Meaning depends on agree- ment, 112, 181, 216, 223, 232, 240, 259, 260 Importance of context, 170, 179. 193. 268 Rough distinction between different (grammatical) sortsof words, 5, 67, 1 73-5, 257 [b. ) Its faults and difficulties : — Discontinuity : — Self-contradiction, how far permissible, 19, 73, 140 »., 183, 227, 243 Clumsiness of descriptive names, 91, 94, 95, 165, 230, 241, 249 Temporary disappearance ofadistinction, 100, 112, 150, 181, 232, 234 Over- conservative tendency : Knowledge progressive, 32, 73, 88«., 94, 123, 182, 221,241-5,248,256 Spoilt words, 88-97, 247 Slavery to words, 12, 20, 74. 87-97. 183. 2'7. 221, 241-4, 269 Need of ' realising our ab- stractions ' : Distinction between things and attributes, 62-7, 85, 97. .141. 257-9 Assertions of ' tendency,' 1 20-4, 129, 249 The objection 'truism, or untrue,' 125, 127, 207, 265 275 DISTINCTION LAN Language and its Difficul- ties (cont.^ : — (c. ) How the difficulties are met in practice : — Tlie epithet ' so-called,' 20, 59 «., 182, 243, 260, 261 Named unrealities, 14, 62-7, 85, 141, 24s, 253, 257-9 'Reference-names,' 58 «., 179-90, 215, 238 Symbolic function of words, 189 Laws of thought, in application, 56, 71, 109, 117, 217. 230 Leibnitz, 266 Liberal and Conservative, 45, 98, 241 Life and death, 23 n., 165 Locke, 264-6 Logic, 21, 56, 147, 15s, 190, 191, 198, 201, 267 Major premiss, 190, 267 Man and beast, 29, 77, 140 »., 146, 161, 261 — and child, 21, 83, 169, 198 n. Matter, 64, 72, 258 Meaning depends on distinction, IS. 27, 99-112, 140, 19s, 198, 231 — and value, 15, loi, 231, 253 — destruction of, 99-112, 129, 163, 195, 222, 231, 254, 260 — predicative, 99 — requires agreement, 112, 181, 216, 223, 232, 240, 259, 260 Melting snow, &c. , 20 Metaphor, 151 Metaphysics, 7, 54, 58, 74, 209, 219-22, 257-9, 265 Middle term, 190, 267 Midway alternative neglected, 198-200 PAR Mill, J. S., 99, 108, 128, 261-3, Minor premiss, 231 «., 268 Moore, Rev. A., 161 Nameable thing, 14 «., 60-9, 102, 245, 253-9 Named unrealities, 14, 62-7, 85, 141 ; and see above Names, abstract, 49, 62, 69, 85, 86, 175, 24s, 257-9 — general, 14, 24, 55, 94, 168- 78, 180, 182, 220,223,242-8, 253-6 — invention of, 170, 255 — need for rough, 8g, 95, 106, 242, 247, 259 — proper, 168-72, 179-84 — symbolic function of, 189 — used for reference, 58 n., 179- 90, 215, 238 Narrowness of view, 98, 134, 199 Nature, continuity of, 71-9, 82, 84, 104, ,198, 259-61 Nature, natural, 102, 210, 216 'Necessary' truth, 213, 219 Neglect of midway alternative, 198-200 Nicknames, 170 «., 171 Occasion, special, 5, 13, 35, 61, 96, 106 «., 117, 118, 131-42, 234 ; and see ' Special Rele- vance ' Opinion, questions of, 41, 155, 194 Organic and inorganic, 23 n. Organism and environment, 75 k. Origin of knowledge, 156 ' Otherness,' 14 ; and see 'Back- ground ' Paradox, 36, 80, 133, 213 Parasites, 75 n. INDEX 277 PAS Passing purpose, 13, 60, 63, 67, 139, 141, 144, 177. 187, 234, 237, 247 Perception and sensation, 153 Permanence and ' thinghood,' 65, 257 Philosophy, science, and common- sense, 10, 30-40, 44, 63, 65, 74, 84, 133, 13s, 138, 147, 225, 233, 250, 258 ' Point,' in names, too »., 102 ; and see ' Meaning ' Population and subsistence, 123 Possibility and probability, 109 ' Practical certainty,' no Predicate terms used for reference, 187-90 Predicative meaning, 99 Producible and real, 66, 72 Progress of knowledge, 32, 73, 88 n., 94, 123, 182, 221, 241-S, 248, 256 'Proper' and 'general' names, 168-72, 179-84 Proverbs, 147 Provisional agreement, 60, 141, 187, 211, 218 Purpose, passing, 13, 60, 63, 67, 139, 141, 144, 177. 187, 234, 237. 247 Question-begging, 182, 185, 186, 188, 189, 213, 216, 217 Question of fact and of opinion, 41. IS5. 194 — of names, 28, 52, 91, 240 — right to waive, 59, 89, 141, 181, 184, 188 Quorum, &c., 68, 69 Quotient, &c., 61 Painbow, Sic, 64 Real and producible, 66, 72 SCI Reality, admitted, 58-70, 85, 112, 131, 141, 187, 213, 227, 24s Reality, and appearance, 215 Reality of attributes, 62-7, 85, 97. 141, 257-9 ' Reference-names, 58 «., 179- 90, 215, 238 Relations and terms, 194, 195 Relativity of all distinctions, 143, 164, 223, 234, 246 Relevance, general and special, 139. 144. 162, 166, 179, 234 ; and see ' Special Occasion ' Religion and science, 182, 185, 189 Resemblance, degree of, 137 — essential, 6, 92, 137, 164, 230 Rival ideals, 35, 41-57, 89, 133, 193, 211, 212, 236, 237, 250 Romanes, G. J., 161 Rough names, need for, 89, 95, 106, 242, 247, 259 Rule and exception, 36, 99, 109, 122, 127, 133, 135, 147, 226 Scepticism and its limits : — ' casts itself out,' 203 The spirit that questions, 203 Its essence is casuistry, 12, 22, 25, 114, 185, 204, 223 Its aim is explanation, 207 Belief and doubt as abstractions, 212, 250 Truth as questionable, 213, 218, 251 ; and see 'Truth and Falsity ' Occasional irrelevance of the sceptical demand, 131-49, 183, 216-24 Doubt has no force unless in- spired by vision of truth, 223, 224 Science, philosophy, and common- sense, 10, 30-40, 44, 63, 65, 74, 278 DISTINCTION SEL 84. 133. I3S. 138, 147. 225. 233, 250, 258 Self-contradiction, how far per- missible? 19, 73, 140 «., 183, 227, 243 Slavery to words, 12, 20, 74, 87- 97, 183, 217, 221, 241-4, 269 - 'So-called,' 20, S9«., 182, 243, 260, 261 Socialism, 159 Socratic method, 24, 25, 104, 1 14- 49, 166, 183, 238 'Special occasion,' J, 13, 35, 61, 96, 106 »., 117, 118, 131-42, 234 ; and see ' Special Rele- vance ' ' Special relevance,' 139, 144, 162, 166, 179, 234 ; and see ' Special Occasion ' Species and variety, 77, 157, 161 Spirit, dungeon, &c., 177 Spoilt words, 88-97, 247 Sqztare, 169, 183, 185 Straight and crooked, 29, 75, 106, 109, 213, 261 Stress on a distinction, 145, 164, 180, 181, 217 Substantive, verb, and adjective, s. 67, 173-S. 257 Suez and Panama canals, 137 Symbolic function of words, 189 Tact, common-sense, 9, 17, 30-40, 63, 83, 93, 94, 106 «., 135. 147, 225, 228 Tastes, opposition of, 42, 239 Tendency, assertions of, 120-4, 129, 249 Terms and relations, 194, 195 Theory and dogma (or assump- tion), 187-9 — and fact, 41, 154, 194, 206 — and practice, 31, 44, 74, 107, III, 228 VER ' Thinghood,' 14, 58-70, 85, 112, 244, 24s, 253, 257-9 Things and attributes, 62-7, 85, 97, 141. 257-9 — immaterial, 60-9, 85, 97, 229, 245, 253, 257 — 'nameable,' 14 «., 60-9, 102, 245. 253-9 Thought and language, 14, 17, 21, 99, IS7. 195. 221, 241-9, 255. 256 Trough of cyclone, 61 Truism, 82, 125, 127, 197, 207, 213, 265 Truth and error mingled, 24, 34, 35, 81, 85, 206, 219, 221, 222, 239 Truth and falsity, 23, 28, 76, 85 , 105, 207, 219, 222, 245, 254 Truth, 'necessary,' 213, 219 Uniformity of causation, 127, 128, 208 Unreal distinctness, 22, 88 Unrealities, named, 14, 62-7, 85, 141, 245, 253, 257-9 Unreality, 58-70 ; and see ' Reality ' and ' Nameable Things ' Use of distinction for argument, 144, 164, 217 Utility (Caimes), 124 Vagueness, excuses for, 47 ?;., 89, 94, 106, 117, 121, 132, 143, 147, 183, 201, 222, 233, 238, 242, 247 'Validity' of distinctions, 143, 234 'Value' and meaning, 15, loi, ' 231 ; and see ' Background ' Verb, substantive, and adjective, s. 67. 173-5. 257 INDEX 279 VOL Voluntary and involuntary, 153 Waiving a question, 59, 89, 141, 118, 184, 188 Words as instruments of ex- pression, 12, 25, 69, 177, 179, 215, 240, 242, 244-7, 253-6. 269 WOR Words as ' signs of ideas,' 246 — clumsiness of, 91, 94, 95, 165, 230, 241, 249 — slavery to, 12, 20, 74, 87- 97, 183, 217, 221, 241-4, 269 — spoilt, 88-97, 247 — symbolic function of, 189 PRINTED BY SrOTTISWOODE AND CO., NEW-STREET SQUARE LONDON A CATALOGUE OF WORKS GENERAL LITERATURE PUBLISHED BY MESSRS. LONGMANS, GREEN. & CO., 89 PATERNOSTER ROW, LONDON, EX. MESSRS. LONGMANS, GREEN, & CO. Issue tht undermentioned Lists of their Publications, which may be had post free on application : — I. Monthly List of New Works AND New Editions. Quarterly List of Announce- ments AND New Works. Notes on Books; being an Analysis of the Works published during each Quarter. Catalogue of Scientific Works. Catalogue of Medical and Surgical Works. 6. Catalogue of School Books AND Educational Works. Catalogue of Books for Ele- mentary Schools and Pupil Teachers. Catalogue op Theological Works by Divines and Mem- bers of the Church of England. Catalogue of Works General Literature. in ABBEY (Bev. C. J.) and OVURTOlf (Rev. J. H.).— THE ENGLISH CHURCH IN THE EIGHTEENTH CENTURY. Cr. 8vo. 7s. 6d. ABBOTT (Evelyn).— A HISTORY OF GREECE. In Two Parts. ' Part I. — From the Earliest Times to the Ionian Revolt. Cr. 8vo. lar. 6d. Part n. Vol. L— 500-445 B.C. [In the Press.'} Vol. II.— ^In Preparation.] HELLENICA. A Collection of Essays on Greek Poetry, Philosophy, History, and Religion. Edited by Evelyn Abbott. 8vo. i6s. /y^ MyUKSSH (A. H. Dyke) and KAITSOME (Cyril).— A HANDBOOK IN OUTLINE OF THE POLITICAL HISTORY OF ENGLAND TO 1S90. Chronologically Arranged. Crown gyp. df. ACTOMllirBa).- 4;. td. -MODERN COOKERY. With 150 Woodcuts, Fcp. 8vo, 2 > A CATALOGUE OF BOOKS IN GENERAL LITERATURE A. K. H. B.— THE ESSAYS AND CONTRIBUTIONS OF. Crown 8vo. 3J. 6d. each. Autumn Holidays ofa Country Parson. Changed Aspects of Unchanged Truths. Commonplace Philosopher. Counsel and Comfort (Vom a City Pulpit. Critical Essays of a Country Parson. East Coast Days and Memories. Graver Thoughts of a Country Parson. Three Series. , Landscapes, Churches, and Moralities, Leisure Hours in Tonn. Lessons of middle Age. Our Little Life. Two Series. Our Homely Comedy and TVagedy. Present Day Thoughts. Becreations of a Country Parson, Three Series. Seaside Musings. Sunday Afternoons in the Parish Church of a Scottish University ,, City. , ,- , ' To Meet the Dky ' through the Christian Year ; being a Text of Scrip- ture, with an Original Meditation and a Short Selection in Verse for Every Day. Crown 8vo. ^s. 6d. ^ — TWENTY-FIVE YEARS OF ST. ANDREWS. 1865-1890. 2 vols. 8vo. Vol. I. 8vo. 12S. [Ready."] Vol. II. [In prefaration.'] AMERICAN WHIST, Illustrated :. containing the Laws and Principles of the Game, the Analysis of the New Play. By G. W. P. Fcp. 8vo. 6s. 6d. AMOS (Sheldon).— A PRIMER OF THE ENGLISH CONSTITUTION AND GOVERNMENT. Crown 8vo. 6s. ANNUAL REGISTER (The). A Review of Public Events at Home and Abroad,' for the year 1890. 8vo. i8j. *,* Volumes of the 'Annual Register' for the years 1863-1889 can still beiad. AITSTEY (F.).— THE BLACK POODLE, and other Stories. Crowh 8vo. 2.!. boards, ; zs. 6d. cloth. ' VOCES POPULI. Reprinted from PuncA. First Series, with 20 Illus- trations by J. Bernard Partridge. Fcp. 4to. 5^- ARISTOTLE— The Works of. THE POLITICS, G. Bekker's Greek Text of Books I. HI.. IV. {VII. ), with an English Translation by W. E. Bolland, and short Introductory Essays by Andrew Lang. Crown 8vo. ys. 6d. THE POLITICS, Introductory Essays. By Andrew Lang. (Fyom Bolland and Lang's ' Politics '. ) ^ Crown 8vo. 2s. 6d. THE ETHICS, Greek Text, illustrated with Essays and Notes. By S|r Alexander Grant, Bart. 2 vols. 8vo. 316/. THE NICOMACHEAN ETHICS, newly translated into English. By Robert Williams. Crown 8vo. js. 6d. ABMSTRONG (Ed.),— ELISABETH FA,RNESE : the Termagant of Spain. ' ^ V , ARMSTEOITG (G. F. Savage-),— POEMS : Lyrical and Dramatic, Fcp. 8yo. 6s, ^ By the Same Author. Fcp. 8vo, King Saul. tf. King David, 6s. King Solomon, 6;. Ugone ; a Tragedy. 6s. A Garland from Greece, . Poems. 9J. Stories of Wicklow. Poems. 9^. ARMSTRONG (E. J.).^POETICAL WORKS. Fcp.'Svo. «. ESSAYS AND SKETCHES. Fcp, 8vo. is. Mephistopheles In Broadcloth ; a Sa- tire. 4r. One in the Infinite ; a Poem. Crown 8vo. 7J. 6d, The Life and Letters of Edmond J. Armstrong, -js. 6d. PVBUSHED BY M&SSRS. LONGMANS, GREB.N, &• CO. 3 AKNOIiD (Sir Edwin).— THE LIGHT O?- THE WORLD, or the Great Consummation. A Poem. Crown 8vo. 7s. 6d. net. SEAS AND LANDS. With 71 Illustrations. Croivn 8vo. js. 6d. ARirOIil) (Dc T.)-— INTRODUCTORY LECTURES ON MODERN HISTORY. 8vo. ys. 6d. MISCELLANEOUS WORKS. 8vo. 7s. (4. ASHIiET (J. W.).— ENGLISH ECONOMIC HISTORY AND THEORY. Part I. — ^The Middle Ages. Crown 8vo. y. ATEIiIBE. (The) du Lys ; or, An Art Student in the Reign of Terror. By the Author of ' Mademoiselle Mori '. Crown 8vo. 2J. 6rf. , By the Same Author. Crown zs. 6d. each. > MADEMOISELLE MORI. A CHILD OF THE REVOLU- THAT CHILD. TION. , UNDER A CLOUD. ; HESTER'S VENTURE. THE FIDDLER OF LUGAU. ' IN THE OLDEN TIME. BACOU".— COMPLETE WORKS. Edited by R. L. Ellis, J. SpedDino, and D. D. Heath. 7 vols. 8vo. ^^3 ijr. td. ' _ LETTERS AND LIFE, INCLUDING ALL, HIS OCCASIONAL . WORKS. Edited by J. Spedding. 7 vols. 8vo. £/^ ^i. THE ESSAYS ; with Annotations. By Archbishots WhaTELY. 8vo. JOS. 6d. — THE ESS.AYS ; with Introduction, Notes, and Inde* By E. A. Abbott. 2 vols. Fcp. 8vo. 6s. Text and Index only. Fcp. 8vo. zf. 6d. BADMINTON LIBRARY (The), edited by the DuKE op BSAUfort, , assisted by Alfred E. T. Watson. HUNTING. By the DoEE OP Beaufort, and Mowbray Morris. With 53 Illustrations. Crown 8vo. los. 6d. FISHING. By H. Cholmondeley-Pennell. Vol. I. Salmon, Troiit, and Grayling. 158 Illustrations. Crown 8vo. loj. 6d. Vol. II. Pike and other Coarse Fish. 132 Illustrations. Crown 8vD. 10s. 6d. RACING AND STEEPLECHASING. By the Earl of Suffolk and Berkshire, W. G. Craven, &c. 56 Illiistrations. Crown 8vo. 10s. 6d. , SHOOTING. By Lord Walsingh am, and Sir Ralph Payne-Gallwey, Bart. Vol. I. Field and Covert. With 105 Illustrations. Crown 8vo. loi 6ii. Vol. II. Moor and Marsh, With 65 Illustrations, Crown 8vq. ioj. 6d. CYCLING. By Viscount Bury (Earl of Albemarle) and G. Lacy HiLLtEKl With 89 Illustrations. Cirown 8vo./ioj. 61?. ATHLETICS AND FOOTBALL. By Montague Shearman. With 41' Illus- trations. Crown 8v6. ic«. 6d. BOATING. By W. B. Woodgate. With 49 Illustrations. Crown 8vo. los. 6d. CRICKET. ByA.G. Steel and the Hon. R. H.LYTTELTpN. With 63 Illus- trations. ., Crown 8vo. loi. 6d. DRIvfNG. By the DuKE OF BEAUFORT. With 65 IUustra,tions. Crown 8vo. loj. 6d. / [Continued. 4 A CATALOGUE OF BOOKS IN GENERAL LITERATURE BADMinSTTON" LIBRARY Crh.e,)— {continued). FENCING, BOXING, AND WRESTLING. By Walter H. Pollock, F. C. Grove, C. Prevost, E. B. Michell, an(l Walter Armstrong. With 42 Illustrations. Crown 8vo. ioj. 6d, GOLF. By Horace Hutchihson, the Rt. Hon. A. J. Balfour, J^.P,, Andrew Lang, Sir W. G. Simpson, Bart., &c. With 88 Illustrations. Crown Bvo. los. 6d. TENNIS, LAWN TENNIS, RACKETS, AND FIVES. By J. M. and C. G. Heathcote, E. O. Pleydell-Bouverie, and A. C. Ainger. With 79 ' Illustrations. Crown 8vo. loj. 6d. 1 RIDING AND POLO. By Captain Robert Weir, Riding-Master, R.H.G., J. MbRAY Brown, &c. With 59 Illustrations. Crown 8vo.- lor. 6d. SKATING, CURLING, TOBOGGANING, &c. By J. M. Heathcote, C. G. Tebeutt, T. Maxwell Witham, the Rev. John Kerr, Oemond Hake, and Colonel Buck. With 284 Illustrations. Crown 8vo. los. 6d. BAGEHOT ("Walter).— BIOGRAPHICAL STUDIES. 8vo. izs. ECONOMIC STUDIES. 8vo. loj. 6d. LITERARY STUOlES. 2 vols. 8vo. zZs. THE POSTULATES OF ENGLISH POLITICAL ECONOMY. Crown Bvo. ar. 6d. BAGWIBLI. (Richard).— IRELAND UNDER THE TUDORS. {3 vols.) Vols. I. and II. From the first invasion' of the Northmen to the year 1578. 8vo. 325. Vol. III. 1578-1603. 8vo. i8j. BAIIT (Alex.).— MENTAL AND MORAL SCIENCE. Crown Bvo. 10s. 6d. SENSES AND THE INTELLECT. 8vo. 15^. EMOTIONS AND THE WILL. Bvo. 15^. LOGIC, DEDUCTIVE AND INDUCTIVE. Part I., Deduction, 45. Part II., Induction, 6s. 6d. PRACTICAL 'ESSAYS. CroVn! Bvo. 2S. BAKER.— EIGHT YEARS IN CEYLON. With 6 Illustrations. Crown 8vo. 3s. 6d. THE RIFLE AND THE HOUND IN CEYLON. With 6 Illustra- tions. Crown 8vo. 3s. 6d. RAT.T. (The Rt. Hon. T. J.).— THE REFORMED CHURCH OF IRE- LAND (1537-1889). Bvo. 7s. 6d. HISTORICAL REVIEW OF THE LEGISLATIVE SYSTEMS OPERATIVE IN IRELAND (1172-1800). Bvo. 6s. BARIHG-aOTJIiD (Rev. S.).— CURIOUS MYTJIS OF THE MIDDLE AGES. Crown Bvo. 3s. 6d. BEACONSFIELD (The Earl of).— NOVELS AND TAL^. The Hughenden Edition. With 2 Portraits and 11 Vignettes. 11 vols. Crown Bvo. 42s. Endymlen. Lothair. Conlngsby. Tfincred. Sybil. NOVELS AND TALES. Cheap Edition. 11 vols. Crown Bvo. is. each, boards; is. 6d. each, cloth. Yenetia. Henrietta Temple. Contarlni Fleming, &g. Alroy, Ixion, &c. The Young Dnke, te. Ylvian Grey. PUfiLlSHEt) BV MESSRS. LONGMANS, GRE&N, 6* CO. 5 BIICKER {Professor).r-GALLUS ; or, Roman Scenes in the Time of Augus- tus. Illustrated. Post 8vo. 7J. 6A '' CHARICLES ; or, Illustrations of the Private Life of the Ancient Greeks. Illustrated. Post 8vo. js. 6d. BSLIi (Mrs. Hugh).— WILL O' THE WISP : a Story. Crown 8vo. 31. 6d. CHAMBER COMEDIES. Crown 8vo. di. NURSERY COMEDIES. Fop. 8vo. is. 6d. BIiAKE (J.).— TABLES FOR THE CONVERSION OF 5 PER CENT. IN- TEREST FROM A TO 7 PER CENT. 8vo. lar. 6d. BOOK (THE) OF WEDDING DAYS. Arranged on the Plan of a Birthday. Book. With 96 Illustrated Borders, Frontispiece, and Title-page by Walter Crane ; and Quotations for each Day. Compiled and Arranged by K. E^ J. Reid, Mav Ross, and Mabel Bamfield. 4to. 21.1. BBASSET (Lady).— A VOYAGE IN THE ' SUNBEAM,' OUR HOME ON THE OCEAN FOR ELEVEN MONTHS. Library Edition. With 8 Maps and Charts, and 118 Illustrations, 8V0..21J. Cabinet Edition. With Map and 66 Illustrations, Crown 8vo.' 7s. id. ' Silver Library ' Edition. - With 66 Illustrations, Crown 8vo. y. 6d. School Edition. With 37 Illustrations, Fcp. 2j. cloth, or jr. white parchment. Popular Edition. With 60 Illustrations, 410. 6d. sewed, is. cloth. SUNSHINE AND STORM IN THE EAST. Library Edition. With 2 Maps and 114 Illustrations, 8vo. 2ii. Cabinet Edition. With 2 Maps and 114 Illustrations, Crown 8vo. js. dd. Popular Edition. With 103 Illustrations, 4to. 6d. sewed, u. cloth. IN THE TRADES, THE TROPICS, AND THE ■ ROARING FORTIES'. Cabinet Edition. With Map and 220 Illustrations, Crown 8vo. js. 6d. Popular Edition. With 183 Illustrations, 4to. 6d: sewed, u. cloth. THE LAST VOYAGE TO INDIA AND AUSTRALIA IN THE ' SUNBEAM '. With Charts and Maps, and 40 Illustrations in Monotone {20 full-page), and nearly 200 Illustrations in the Text. 8vo. 21J. ■ THREE VOYAGES IN THE ' SUNBEAM '. Popular Edition, With 346 Illustrations, 4to. zs, 6d. BBAY (Charles).— THE PHILOSOPHY OF NECESSITY ; or, La)v in Mind as in Matter. Crown 8vo. ss. BRIGHT (Rev. J. Franck).— A HISTORY OF ENGLAND. 4 vols. Cr. 8vo. Period 1.,-Mediseval Monarchy : The Departure of the Romans to Richard III. From a.d. 449 to 1485. 4s. 6d. Period II.— Personal Monarchy : Henry VII. to James II. From 1485 to 1688. sj. Period III. — Constitutional Monarchy : WiUiam and Mary to William IV. From 1689 to 1837. 7s. 6d. , Period IV.— The Growth of Democracy : Victoria. From 1837 to 1880. 6s. BBTDEIT (H. A.).— KLOOF AND KARROO : Sport, Legend, and Natural / History JB Cape Colony. With 17 Illustrations. 8wo. los. 6d. BtrCKLE (Henry Thomas).- HISTORY OF CIVILISATION IN ENG- LAND AND FRANCE, SPAIN AND SCOTLAND. 3 vols. Cr. 8vo. 24J. 6 A CATALOGUE OP BOO^S IN GENERAL LITERATURE BULL (Thomas).— HINTS TO MOTHERS ON THE MANAGEMENT OF THEIR HEALTH during the Period of Pregnancy. Fcp. -8vo. lA 6d. — THE MATERNAL MANAGEMENT OF CHILDREN IN HEAI-TH AND DISEASE. Fcp. 8vo. u. 6d. BUTLER (Samuel).— EREWHON. Crown 8vo. ^. THE FAIR HAVEN. A Work in Defence of the Mfraculous Element in our Lord's Ministry. Crown Svo. js, 6d. LIFE AND HABIT. An Essay after a Completer View of Evolulion. Cr. Svo. 7J. 6d. fiVOLUTION, OLD AND NEW. Ctowh Svo. loj. 6d ' ' UNCONSCIOUS MEMORY. Crown Svo. js. U. ALPS AND SANCTUARIES OF PIEDMONT ANfi THE CANTON TICINO. Illustrated. Pott 4to. xas. 6d. SELECTIONS FROM WORKS. Ctown Svo. 71. 6d. LUCK, OR CUNNING, AS THE MAIN MEANS OF ORGANIC MODIFICATION? Crown Svo. 71. 6d. EX VOTO. An Account of the Sacro Monte or New Jerusalem at" Varallo-Sesia. Crown Svo. loj. 6d. HOLBEIN'S 'LA DANSE'. 3s. CABLYLE (Thomas).— THOMAS CARLYLE: a History of his Life, By J. A. Fkoude. 179S-1835, 2 vob. Cr. Svo. 7s. 1S34-18S1, zvols. Cr. Svo.yj. CASE (Thomas).— PHYSl'CAL REALISM :, being an Analytical Philosophy from the Physical Objects of Science to the Pliysical Data of Sense. Svo. 15*. CHETWYND (Sir George).— RACING REMINISCENCES AND EX- PERIENCES OF THE TURF, a vols. Svo. sis. CHILD (Gilbert W.).— CHURCH AND STATE UNDER THE TtJDORS. Svo. 15s. CHISHOXiM (G, G.).— HANDBOOK OF COMMERCTAL GEOGRAPHY. With 29 Maps. Svo. i6j. CHUBCH (Sir Kiehardy.— Commander-in-Chief of the Greeks in the War of Independence : a Memoir. By Stanley Lane-Poole. Svo. '. ^s. CLrVE (Mrs. Archer).— POEMS. Including the IX. Poemsl Fcp. Svo. 6j. CLODD (Edward).— THE STORY OF CREATION : a Plain Account ql Evolution. With 77 Illustrations. Crown Svo. 33. 6d, CLtTTTERBUCK (W. J.).— THE SHIPPER IN ARCTIC SEAa With 39 Illustrations. Crown Svo. ioj. 6d. ^ ABOUT CEYLON AND BORNEO : being an Account of T\»o Visits to Ceylon, one to Borneo, and How we Fell Out on our Homewajd Journey. With 47 Illustrations. Crown Svo. COLEirSO (J. W.).— THE PENTATEUCH AND BOOK OF JOSHUA CRITICALLY EXAMINED. Crown Svo. 6s. COMYM" (L. If.).— ATHERSTONE PRIORY : a Tale. Crown Svo. as'. 6d. COHnrGTOIT (John).— the ^NEID OF VIRGIL. Translated iiito English Verse. Crown Svo. 6s. THE POEMS OF VIRGIL. Translated into English Prose. Cr. Svo. 6s. COX (Bev. Sir G. ■W.).-^A HISTORY OF GREECE, from the Earliest Period t the Cloister and the Forest ip the Days of the Barons' Wars. Brain Fltz-Coimt, A Story of W^l- lingford Castle and Dorchester Abbpy. HISTORY OF THE CHURCH UNDER THE ROMAN EMPIRE, A.D. 30-476. Crown 8vo. js. 6d. ' CREIQHTOIT (MandeU, D.D.)— HISTORY OF THE PAPACY DUR- ING THE REFORMATIpN. 8vo. Vols. L and II., 1378-1464, sjy. ; Vols. III. and IV., 1464-1518, 247. CRUMP (A.).— A SHORT ENQUIRY INTO THE FORMATION OF POLITICAL OPINION, from the Reign of the Great Families to the Advent of Democracy. 8vo. js. 6d. AN INVE.STIGATION INTO THE CAUSES OF THE GREAT FALL IN PRICES ' which took plflce coincidei»tly with the Demonetjsation of Silver by Germany. 8vo. 6s, . . DABTTE.— LA COMMEDIA DI, DANTp. A New Text, carefully Revised with the aid of the most recent Editions and Collations. Small 8vo. 6j. DE LA SAITSSAYE (Prof. Chantepie).— A WtANUAL OF THE ' SCIENCE OF RELIGION. Translated by Mrs. Colyee Fergusson (?iie Max Muller). Crown 8vo. izs. 6d, DEIiAMD (Mrs.).— JOHN WARD, PREACHER. Cr. 8vo. zs. bds. , as. 6d. cl. SIDNEY : a Novel. Crown 8vo. 6s. THE OLD GARDEN, and other Verses. Fcp. 8vo. 5^- DE REDCIiIPI'E.— THE LIFE OF THE RIGHT HON. STRATFORD CANNING . VISCOUNT STRATFORD DE REDCLIFFE. By ST.\NLEy XvANE-PooLE. With 3 Portraits. Crown 8vo. 7J. 6d. DE SAIiIS (Mrs.).— Works by :— Cakes and Confections a la Mode. Fcp. 8vo. Ts. 6d. Dressed Game and Poultry it la Hade. Fcp. 8vo. IS. 6i. Dressed VegetableB ik la Hode. Fcp. 8vo. xs. 6d. Drinks i. la Kode. Fcp. 8vo. u. 6d. Entries & la Hod^. Fcp. zs. 8vo. 63. Floral Decorations. Fcp. 8vo. ts. 6d. Oysters a la Hode. Fcp. 8vo. xs. 6d. Puddings and Pastry & la Hode. Fcp. 8vo. IS. 6d. Savouries ik la- Hode. Fcp. 8vo. is. 6d. Soups and Dressed Fish a la Mode. Fop. 8vo. xs. 6d. 'Siareets and Supper Dishes k la Hode. Fcp. 8vo. IJ. 6d. Tempting Dishes for Small Incomes. Fcp. 8vo. ij. 6d. Wrinkle;) and Notions for every House- hold. Crown 8vo. xs. 6d. DE TOCQUJSVILLE (Alexis).— DEMOCRACY IN AMERICA Trans- lated by Hehry Reeve, C.B. 2 vols. Crown 8vo. x6s. DOTTGAIjIi (Ii.).— BEGGARS ALL ; a Novel. Crown 8vo. 6s. DOWELL (Steplien).-A HISTORY OF TAXATION AND TAXES IN ENGLAND. 4 vols. 8vo. Vols. I. and II., The History of Taxation, 2xs. Vols. III. and IV., The History of Taxes, 21J. DOYtiE (A. Conan).— MICAH CLARKE : a Tale of Monmouth's Rebellion. With Frontispiece and Vignette. Crown 8vo. 31. 6 CO. 9 FOX (C. J.).— THE EARLY HISTORY OF CHARLES JAMES FOX. By the Right Hon. Sir. G. O. Teevelyan, Bart. Library Edition. 8vo. iSj. | Cabinet Edition. Crown Svo. ts. FBAirCIS (Francis).— A BOOK ON ANGLING: including full Illustrated Lists of Salmon Flies. Post Svo. 15.5. PBEEMAM" (is. A.).— THE HISTORICAL GEOGRAPHY OF EUROPE. With 63 Maps. 2 vols. Svo. 31J. dd. FROtTDE (James A.).— THE HISTORY OF ENGLAND, from the Fall of Wolsey to the Defeat of the Spanish Armada. 12 vols. Crov^n Svo. £z 2J. THE DIVORCE OF CATHERINE OF ARAGON : The Story as told by the Imperial Ambassadors resident at the Court of Henry VIII. in Usum Laicorum. Svo. i&r. THE ENGLISH IN IRELAND IN THE EIGHTEENTH CEN- TURY. 3 vols. Crown Svo. i8j. '■ — SHORT STUDIES ON GREAT SUBJECTS. Cabinet Edition. 4 vols. Cr. Svo. 245. | Cheap Edit. 4 vols. Cr. Svo. y. 6d. ea. ^ CiESAR : a Sketch. Crown Svo. ■=fi. dd. OCEANA ; OK, ENGLAND AND HER COLONIES. With 9 Illus- trations. Crown Svo. 2.r. boards, 2^. ^. cloth. THE ENGLISH IN THE WEST INDIES; or, the Bow of Ulysses. With 9 Illustrations. Crown Svo. 2i. boards, as. dd. cloth. THE TWO CHIEFS OF DUNBOY; an Irish Romance of the Last Centiu^. Crown Svo. y. dd. THOMAS CARLYLE, ii History of his Life. 1795 to 1S3S. 2 vols. Crown Svo. js. 1834 to iSSi. 2 vols. Crown Svo. ys. G-AIiIiWEY(SirRalplxPayne-).— LETTERS TO YOUNGSHOOTERS. (First Series. > On the Choice and Use of a Gun. Crown Svo. 7J. dd. GAKDIITEE (Samuel Eawson).— HISTORY OF ENGLAND, 1603- 1642. 10 vols. Crown Svo. price ds. each. A HISTORY OF THE GREAT CIVIL WAR, 1642-1649. (3 vols.) Vol. I. 1642-1644. With 24 Maps. Svo. {out qffrint). Vol. II. 1644-1647. With 21 Maps. Svo. 241. Vol. III. 1647-1649. With 8 Maps. s&s. THE STUDENT'S HISTORY OF ENGLAND. Vol. I. B.C. 5S-A.D. 1509, with 173 Illustrations, Crown Svo. 4?. Vol. II. 1509-16S9, with 96 Illustrations. Crown Svo. 41. , Vol. III. 16S9-1SSS, vrith 109 Illustrations. Crown Svo. 41. Complete in i vol. With 378 Illustrations. Crown Svo. izr. - A SCHOOL ATLAS OF ENGLISH HISTORY. A Companion Atlas to ' Student's History of England '. 66 Maps and 22 Plans. Fcap. 4tO: 51, GIBEENE (Agnes).— NIGEL BROWNING. Crown Svo. s^. GOETHE.— FAUST. A New Translation chiefly in Blank Verse ; with Intro- duction and Notes. By James Adey Birds. Crown Svo. ds. FAUST. The Second Part. A New Translation in Verse. By James Adey Birds. Crown Svo. ds. GREEN (T. H.)— TJiE WORKS OF THOMAS HILL GREEN. (3 Vols. ) Vols. I. and II. Svo. xds. each. Vol. III. Svo. 21J. • THE WITNESS OF GOD AND FAITH : Two Lay Sermons. Fcp. Svo. 2f. GBEVILIiE (C, C. F.).— A JOURNAL OF THE REIGNS OF KING* GEORGE IV., KING'WILLIAM IV., AND QUEEN VICTORIA. Edited by H. Reeve. 8 vols. Crown Svo. ds. each. lo A- CATALOGUE Off. BOOKS IN GENERAL LITERATURE GWILT (Joseph).— AN EiNCYCLOP.EDl.^ , 0.1? ARCHITECTURB,, Willi more than 1700 Engravings on Wood. 8vo. 52^. 6n?. HAGGARB (H. Eider).— SHE. With 32 Illustrations. Crown 8vo. 3s. W. ALLAN QUATERMAIN: With 31 Illustrations. Crowii'Svo. 35. (id. MAIWA'S REVENGE. Crown Svo. li. boards, u. bd. doth. COLONEL QUARITCH, V.C. Crown 8vo. ■y, 6rf. CLEOPATRA: With 29 Illustrations. Crown Svo. 3J. 6rf. BEATRICE. Crown 8vo. 3*. 6(/. . , '. ; : ERIC BRIGHTEYES. With 51 Illustrations. Crown 8vo, 6s. HAGGABD (H. Eider) and LAHTG (Andrew).— THE WORLD'S ' DESIRE. Crown 8to.6j. HALF HOURS WITH THE MILLIONAIRES. Edited by B. ' B. West. Crown 8vo. 6/. ^ HAZiLrWEIiIi-PHIIiIiIPPS (J. O.)— A CALENDAROFTHEHALLI- WELL-PHILLIP P.S COLLECTION OF SHAK.ESPEAR EAN RARITIES Second Edition. Enlarged by Ernest E. Balder. 8vo. las. td^ - OUTLINES OF THE LIFE OF SHAKESPEARE. With numerous Illustrations and Facsimiles. 2 vols. Royal 8vo. zis. < HAEEISOW (Jane E.).— MYTHS OF THE ODYSSEY IN ART AND LITERATURE, Illustrated with Outline'Drawings. 8vo. i8j; HAERISOM" (F. Bayford).— THE CONTEMPORARY HISTORY OF THE FRENCH REVOLUTION. Crown 8vo. 3i. ' 61^. HAETE (Bret).— IN THE CARQUINEZ WOODS. Fcp. 8vo. is. bds,, js. 6d. cloth. BY SHORE AND SEDGE. i6mo. is. ON THE FRONTIER. i6mo. is. HAETWIG (Dr.).— THE SEA AND ITS LIVING WONDERS. Witfi i;2 Platesand 303 Woodcuts. 8vo. 7s. net. THE TROPICAL WORLD. With 8 Plates and 172 V^oodouts," 8vo. 7s. nei., THE POLAR WORLD. With 3 Maps, 8 Plates and 85 Woodcuts. 8vo. 7s. net. THE SUBTERRANEAN WORLD. With 3 Maps and 80 Woodcuts. 8vo.7s.net. THE AERIAL WORLD. With Map, 8 Plates and 60 Woodcuts. 8vo.,7j. net. HAVELOCK.— MEMOIRS OF SIR liENRY HAVELOCK, K.C.B. By John CLARfc Masshman. Crpwin Syo. 3^. 61^. . HEAEU (W. Edward).— THE GOVERNMENT OF ENGLAND: its Structivre and its Development. 8vo. 16s. THE ARYAN HOUSEHOLD : its Structure and ts Development. An Introduction to Comparative Jurisprudence. 'Svo. 16s. HISTOEIC TOWlfS. Edited by E. A. Freeman and Rev. William Hunt. With Maps and Plans. Crown 8vo 31. 6d. each. Bristol. By Rev. W. Hunt. Carlisle. By Dr. Mandell Creighton. Cinque Ports, By Montagu Burrows. Colchester. By Rev. E. L. Cutts. Exeter. By E. A. Freeman. London. By Rev. W. J. Loftie. Oxford. By Rev. C. W. Boase. Winchester. By Rev. G. W. Kitchin. New York. By Theodore Roosevelt. Boston (U.S.). By Henry Cabot Lodge. York. By Rev. Jajnes Raine. [/« Reparation, PUBLISHED BY MESSRS. LONGMANS, GREEN, &> CO. it BEODGSON CSHadworth H.).— TIME; AND SPACE: a Melapliysical Essay. 8vo. lbs. > ' — THE THEORY OF PRACTICE : an Elhif ^1 Enquiry^ 2 vols. 8vo. 24J. =- THE PHILOSOPHY OF REFLECTION. ' 2 vols. 8vo. 2ij. OUTCAST ESSAYS AND VERSE TRA'^NSLATIONS. Crown Svo. 8s. 6d. HOWlTTCVirmiamy.— VISITS TO REMARKABLE PLACES. 80 Illus- trations. Crown Svo. y, 6ct. HULLAH (John).— COURSE OF LECTURES ON THE HISTORY OF MODERN MUSIC. 8vo. 8j. 6*. COURSE OF LECTURES ON THE TRANSITION PEKIOD OF MUSICAL HISTORY. 8vo. loi. &/. HUME.— THE PHILOSOPHICAL WORKS OF DAVID HUME. EiliieU by T. H. 'Green and T. H. Gkose. 4 vols. Svo. 561. HUTCHIITSOIT (Horace).— FAMOUS GOLF LINKS. By Horace G. Hutchinson, Andrew Lang, H. S. C. livt.KAUD, T. RuTHF-KroKii Clark. &c. With numerous Illustrations by F. P. Hopkins, T. Hodges, H. S. King, &c. Crown Svo, 6s. . H0TH (Alfred H.).— THE MARRIAGE OF NEAR KIN, considered with respect to the Law of Nations, the Result of Experience, and the Teacliings of Biology. Royal Svo. ais. HTGEIiO'W (Jean).-^POETICAL WORKS. Vols. I. and II. Fcp. Svo. I2S. Vol. III. Fcp. 8vb. 5i. — LYRICAL AND OTHER POEMS. Selected from the Writings of Jean Ingelow. Fcp. Svo. 2s. 6d. cloth plain, 3s. cloth jji't- VERY YOUNG and QUITE ANOTHER STORY : Two Stories. Crown 8vo. 6s. INGKAM (T. Dunbar).— ENGLAND AND ROME : a History of the Relations between the Papacy and the Enghsh State and Church from the Norman Conquest to the Revolution of 16S8. Svo. i+r. JAMESOM" (Mrs.).— SACRED AND LEGENDARY ART. With 19 Etch- ings and 187 Woodcuts. 2 vols. Svo. zos. net. . l^EGENDS OF THE MADONNA, the Virgin Mary as represented in Sacred and Legendary Art. With 27 Etchings and 165 Woodcuts. Svo. loj. wf/. LEGENDS OF'THE MONASTIC ORDERS. With 11 Etchings and 88 Woodcuts. Svo. lOi. net. HISTORY OF OUR LORD. His Types and Precursors. Corapleled by Lady EiASTLAKE. With 31 Etchings and 281 Woodcuts. 2 V0I5. Svo. 20i. nel. JEPPEBIES (Kichard).— FIELD AND HEDGEROW. Last Essays. Crown Svo. y. 6d. XHE STORY OF MY HEART : My Autobiography. Crown Svo. js. 6(1. r£D deer. With 17 Illustrations by J. Charlton and H. Tunaly, Crown Svo. js. 6d. JEiraniTGS (Rev. A. C.).^ECCLESIA ANGLICANA A History of the Church of Christ in England. CVown Svo. 7*. bd. 12 A CATALOOUE OF BOOKS IN GENERAL LITERATURE JOHITSOM" (J. & J. H.).— THE PATENTEES MANUAL; a XreaUsstOS . the Law and Practice of Letters Patent. , 8vq. ipj. 6rf. ,. ,. JORDAlfCWilliamlieighton).— THESTANDARD6fvaltJe. 8vo.6f. JUSTIWIAIf.— THE INSTITUTES OF JUSTINIAN ; LalSn Text, with English' Introduction, &c. By Thomas C. Sandars. 8vo. iis. BLAJiISCH (M. M.).— BIBLE STUDIES. Part I. The Prophecite ol Balaam. Svo. loi. 6rf. Part II. The Book of Jonah. Svo. zos.^il. _ KALISCH (M. M.).— COMMENTARY 6iM THE OLD TESTAMEisiT • with a New Translation. Vol. I. Genesis, 8vo. i8j. , or adapted foi- the General Reader, lai. Vol. II. Exodfls, ^iSjr.'.'ot aidaptedfor the Gener§il Rga^i^ i|f^. Vol. III. Leviticus, Part I. 151., or adapted for the General! feeaSef, is. VplJV. Levitiqus, Part II. j$s., or adapted forthe General Reajdec, Ss. , , . KAHT (Immanuel).— CRITIQUE OF PRACTICAL REASON, !A!nD OTHER WORKS ON THE THEORY OF ETHICS. 8vb. 12s. 6S. ■ ' I ' INTRODUCTION TO LOGIC. Translated by T. K. Abbott. Notes by S. T. Coleridge; 8vo. 6r. .KIJ.I1ICK (Rev. A. H.).— HANDBOOK TO MILL'S SYSTEM OF LOGIC. Crown 8vo. 3J. 6d. KNIGHT (E. r.).— THE CRUISE OF THE ' ALERTE' ; the Narratjye gf a Search for Treasure on the Desert Island of Trinidad. With'2 MapS'and 23 Illustrations. Crown 8vo. 3J. 6d. SAVE ME FROM MY FRIENDS : a Novel. CroWn 8vo;, &s. ' ' T.AT>T> (George T.).— ELEMENTS OF PHYSIOLOGICAL PSYCHO- LOGY. Bvo. 2IJ. ,, , ,, OUTLINES OF PHYSIOLOGICAL PSYCHOLOGY. A Text-Book of Mental Science for Academies and Colleges. 8vo. lar. 1- 1 LAITG (Andrew).— CUSTOM AND MYTH: Studies of Early Usage aivd Belief. With 15 Illustrations. Crown 8vo. ^s. 6d. BOOKS AND BOOKMEN. With 2 Coloured Plates and 17 Illustra- tions. Fcp. 8vo. zs. 6d. net. — LETTERS TO DEAD AUTHORS. Fcp. 8vo.' zs. 6d. net. OLD FRIENDS. Fcp. 8vo. zs. 6d. net. LETTERS ON LITERATURE. Fcp. 8vo. 2J. 6rf. ««<. '' ', ' BALLADS OF BOOKS. Edited by Andrew Lang. Fcp. Bvo. 'f>s. THE BLUE FAIRY BOOK. Edited by Andrew LANS. With 8 Plates and 130 Illustrations in the Text. Crown 8vo. 6s. THE RED FAIRY BOOK. Edited by Andrew Lang. With 4 Plates and 96 Illustrations in the Text. Crown Svo. ds. - THE BLUE poetry BOOK. With 12 Plates and 88 Illustrations in the Text. Crown Svo. ds. ANGLING SKETCHES. With lUustrations by W. G. BURN- MURDOCH. Crown Svo. ^s. fid. LAVIS8E (Ernest).— GENERAL VIEW OF THE POLITICAL' MlS TORY OF EUROPE. Crown Svo. 51. PUBUSHBD BY MESSRS., l-Oi^MANSi GMSA^^ &• C0> 13 LAYAKD (ITina P.).— POEMS. Growp 8vo. fesj i, li >): i, > hn, IiECKY(W. B. H.)J— HISTORY OF ENGLAND IN THE EIGHTEENTH • CENTURY. Library Edition, 8vp. Vols. I. and 11. 1700-1760. , 36*. Vols. Hi. and IV. 1760-1784. 36^. Vols. V. and VI. 1784-1793.^36^. Vols., VII. and VIII. i?79a-i8oQ. 361. Cabinet Edition, 12 vols. Croivn tivo. 6s. eacb. \Jn,eciufi^e of Publication in Monthly Volumes. THE HISTORY OF EUROPEAN MORALS FROM AUGUSTUS TO CHARLEMAGNE. 2 vols.- CMwn 8h^o. i6j. . . . : , HISTORY OF THE RISE AND INFLUENCE OF THE SPIRIT OFRy^JlONALISM IN EUROPE, 2 vols. Crown 8vo. i6i. " ' TOEMS. Fcap. 8vo. ,5J,; '' ' ' ' < ■•, LEES (Jl A.) and CIjUTTEBBTJCK (W. J.).— B.C 1887, A RAMBLE IN BRITISH COLUMBIA. With Map and 7^ lUusts. Cr. 8vo. y. 6d. LEWES (George Henry).— THE HISTORY OF PHILOSOPHY, from - ,Thate^ to Comte, 2 vo s. 8vo. 32^. LIDDisLL (Colonel K. T^.— MEMOIRS OF THE TENTH ROYAL HUSSARS. With Nuiperous Illustrations.. , 2 vols. Imperial 8vo. 631. LLOYD (P. J.).— THE SCIENCE OF AGRICULTURE. 8vo. i2j. LONGMAM" (Frederick W.).— CHESS OPENINGS. Fcp. 8vo. zs. 6(i. — : FREDERICK THE GREAT AND THE SEVEN YEARS' WAR. Fcp. 8yo. as. 6d. LOITGMOBE (Sir T.)— RICHARD WISEMAN, Surgeon and Sergeant- Surgeon to Charles II.' A Biographical Study. With Portrait. 8vo.ioj.6rf. LOimOM' (J. C.).— ENCYCLOP.iEDIA OF GARDENING. With 1000 Woodcuts. 8vo. 2ii. _ ENCYCLOPAEDIA OF AGRICULTURE ; the Laying-out, Improve- ment, and Management of Landed Property. With iioo Woodcuts. 8vo. 2if. ENCYCLOP.iEDlA OF PLANTS; the Specific Character, &o., of all Plants found in Great Britain. With 12,000 Woodcuts. 8vo. 42J. LUBBOCK (Sir J.).— THE ORIGIN OF CIVILISATION and the Primitive Condition of Man. With 5 Plates and 20 Illustrations in the Text. 8vo. lis. T.-VAT.T. (Edna).— THE AUTOBIOGRAPHY OF A SLANDER. Fcp. 8vo. IS. sewed. LYUE (Lionel W.).— AN INTRODUCTION TO ANCIENT HISTORY. With 3 Coloured Maps. Crown 8vo. y. JSIACAULAY (Lord).— COMPLETE WORKS OF LORD MACAULAY. Library Edition, 8 vols. 8vo. ;^s JJ. I Cabinet Edition, r6 vols, post 8vo. ■ I £ii6s. '■ HISTORY OF ENGLAND FROM THE ACCESSION OF JAMES THE SECOND. Popular Edition, 2 vols. Crown 8vo. 51. I People's Edition, 4vols. Crown 8vo.i6j. Student's Edition, 2 vols. Crown 8vo. Cabinet Edition, 8 vols. Post 8vo. 48J. I2J. ' Library Edition, s vols. 8vo. £/[. CRITICAL AND HISTORICAL ESSAYS, WITH LAYS OF ANCIENT ROME, in i volume. Popular Edition, Crown 8vo. zs. 6d. .Authorised Edition, Crown 8vo. 2s. 6d., or jf. 6d. gilt edges. ' Silver Library ' Edition. With Por- trait and Illustrations to the ' Lays '. Crown 8vo. 3s. 6d._ [CoAtimted. 14 A CATALOGUE OF BOOKS IN GENERAL LITSRA TURE MACAULAY (Lord).^ESSAYS (wnftwKsifl. CRITICAL AND HISTORICAL ESSAYS. , , - Student's Edition. Crown 8vo.6j. | TrevelyanEdition,2vals.Crpwn8vo.9i. People's Edition, 2 vols. Crown 8vo. Zs. Cabinet Edition, 4 vols. Post 8v6. Z4jr. ' Library Edition, 3 vols. 8vo. 3&.. "' ESSAYS whichmay behadseparately, price6rf. eachsewed. ij. eachtloth. Kanke and Gladstone, > •'' Hilton and Machiavellt. Lord Bacon. Lot'd Clive. Lord Byron, and the Comic Drama- tists ot the Restoration, Addison and Walpoie. Frederic the Great. Groker's Boswell's Johnson. Hallam's Constitutional History. Warren Hastin$s(3^. sewed, 61/. cloth). The Earl of Chatham (Two Essays). The Essay on Warren Hastings, anno- I The Essay on Lord Clive, annotated by tated by S. Hales. Fop. 8vo. xs. bi. \ H.CourthopeBowen. Fcp.8vo.2j.6rf. SPEECHES. People's Edition, Crown 8vo. 3J. 6a!. LAYS OF ANCIENT ROME, &c. Illustrated by G. Scharf. Library Edition. Fcp. 410. loj. td. r Bijou Edition, iSmo. 2j. t>d. gjlt top. I Popular Edition, Fcp. 410. bd sewed, I ij, cloth. 1 Illustrated by J. R. Wegueltn. Crown 8vo. 3r. bd. gilt edges. Annotated Edition, Fcp.Svo. I'r.sewed, Cabinet Edition, Post 8vo. 3^. bd. \ is. bd. cloth. MISCELLANEOUS WRITINGS. People's Edition. Crown 8vo. ^j. bd. | Library Edition, 2 vols. Svg. 2tj. MISCELLANEOUS WRITINGS AND SPEECHES. Popular Edition. Crown 8vo. zs. bd- I Cabinet Edition, Post 8vp. 24.t. Student's Edition. Crown 8vo. bs. \ SELECTIONS FROM THE WRITINGS OjTLORD MACAULAy. Edited, with Notes, by the Right Hon. Sir G. 0. Tr-evelyan. Ctown 8vo. bs. THE LIFE AND LETTERS OF LORD MACAPLAY, % t^p Right Hon. Sir G. O. TkilVElyaN. Popular Edition. Crown. 8vo. zs. bd. I Cabinet Edition, 2 vols. Post ivo.i^s. Student's Edition. Crown 8vo. bs. 1 ' Library Edition, 2 vols. 8vo. ■^hs: ■'■ MACDOIfAIiD (George).— UNSPOKEN SERMONS.. Three Series. Crown 8vo. jr. bd. each. THE MIRACLES OF OUR LORD. Crown 8vo. 3J. bd: A BOOK OF STRIFE, IN THE FORM OF THE DIARY OF AN OLD SOUL : Poems. lamo. bs. MACrABREH" (Sir G. A.).— LECTURES ON HARMONY. 8vo. 12s. ADDRESSES AND LECTURES. Crowii 8vo. bs. bd. MACEAIIi (J. W.).— SELECT EPIGRAMS FROM THE GREEK AN- THOLOGY. With a Revised Text, Introduction, Translation, &c. 8vo. i&i. MACIiEOD (Henry D.).— THE ELEMENTS OF BANKING. Crown 8vo. jr. bd. THE THEORY AND PRACTICE OF BAN KING. Vol. I. 8vo. 12s. , Vol. II. 14!. THE THEORY OF CREDIT.' 8vo. ■^ohl.lNew Editionin the Press]- Vol. II. Part I. 41. bd. ; Vol. II. Part II. los. bd. , PU.BI.J§li£.D BY MES^IiS. LONGMANS, GREEN, 6= CO. 15 McCUliLOCH (J. B.)-— THE DICTIONARY OF COMMEiiCE and Com- mercial Navigation. With 11 Maps ani 30 Charts. 8vo. 63s. MACy INE,( John).-SlXTY-THREE ^EAR?' ANGLING, from the Moun- tain Streamlet to the Mighty Tay. Crown Syo. loj. 6rf. lt[AX>M£SBUBY (The Earl of).— MEMOIRS OF AN EX-MINISTER. Crown, 8vo. 7s. 6d. MAlfNERING (Q. E.).— WITH AXE AND ROPE IN THE NEW ZEALAND ALPS. Illustrated. 8vo. i2j. 6d. MANUALS OP CATHOLIC PHILOSOPHY {S/anyAurst Series). JLofilc. By Richaisd F. Clarke. Crown 8vo.' 5/. First Principle! of Knowledge. By John R'ickaby. Crown 8vo. 5^. ■oral Plillosophy (Etiiics and HatucaU Law). By Joseph Rickaby. Crown 8vo.' 5J. Ceneral Metaphysics. By John Ridka- by. Crown Svo. s-f- Fsycliology. By Michael Maher. Crown Svo. 6s. 6rf. Natural Theology.- By Bernard Boedder. Grown Svo. 6s. 6d. A Manual of Political Economy. ByC. S. Devas. 6s. 6d. MARTINEAU (James).— HOURS OF THOUGHT ON SACRED THINGS. Two Volumes of Sermons. 2 vols." Crowii'Svo. yj. 6d. eScb. ENDEAVOURS AFTER THE CHRISTIAN LIFE. Discourses. Crown Svo. 7j. 6d. HOME PRAYERS. Crown Svo. y. 6d. THE SEAT OF AUTHORITY IN RELIGION. Svo. 14^. ESSAYS, REVIEWS, AND ADDRESSES. 4 vols. Crown Svo. 7J. 6d. each. I. Personal: Political. I III. The9logical: Philosophical. II. Ecclesiastical : Historical. | IV. Academical : Religious; MASON (Agnes).— THE StEPS OF THE SUN : Daily Readings of Prose. i6mo. 3J. td. MATTHEWS (Brander).— A FAMILY TREE, and other Stories. Crown Svo. 6j. , ' PEN AND INK— Selected Papers. Crown Svo. 5^. WITH MY FRIENDS : Tales told in Partnership. Crown Svo. &. MAUTTDEK'S TEEASURIES. Fcp. Svo. 6s. each volume The Treasury of Bible Knowledge. By the Rev. J. AVHE. With 5 Maps, 15 Plates,, and 300 Woodcuts. , Fcp. Svo. 6s. ' The Treasury of Botany. Edited by J. LiNDLEy and T. Moobe. With 274 Woodcuts and 20 Steel Plates. 2 vols. Blbgraphloal Treasury. Treasury of Natural History. With goo Woodcuts. Treasury of Geography. With 7 Maps and 16 Plates. ' Scienttfle and Literary Treasury. Historical Treasury. Treasury of Knowledge. MAX MtJLLEB (P.).— SELECTED ESSAYS ON LANGUAGE, MiYTHOLOGY, AND RELIGION. 2 vols. Crown Svo. i6j. „__ THREE LECTURES ON THE SCIENCE OF LANGUAGE. Cr. Svo. y. THE SCIENCE OF LANGUAGE, founded oil Lectures delivered at the Royal Institution in 1S61 and 1863. 2 vols. Crown Svo. zis. HIBBERT LECTURES ON THE> ORIGIN AND GROWTH OF RELIGION, as illustrated by the Religions of India. Crown Svo. 7J. 6d. ' \Continued. i6 A CATALOGUE OF BOOKS IN GENERAL UTERATURE - - — MAX MttliliES (F.)— INTRODUCTION TO THE SClfiNCE OF RE- LIGION ; FourLectures delivered at the Royal Institution. Crown 8vo. ^s. bd. NATURAL RELIGION. The Gifford Lectures, delivered before the University of Glasgow in 1888. Crown 8vo. loj. dd. PHYSICAL RELIGION. The Gifford Lectures, delivered before the University of Glasgow in 1890. Crown 8vo. lor. bd. ANTHROPOLOGICAL RELIGION : The Gifford Lectures delivered before the University of Glasgow in 1891. Crown 8vo. lar. dd. THE'SCIENCE OF THOUGHT. 8vo. 2U. THREE INTRODUCTORY LECTURES ON THE SCIENCE OF THOUGHT. 8vo. ai. td. BIOGRAPHIES OF WORDS, AND THE HOME OF THE ARYAS. Crown 8vo. 7J. dd. ' A SANSKRIT GRAMMAR FOR BEGINNERS. New and Abridged Edition. By A. A. MacDonell. Crown 8vo. &. MAY (Sir a?]iomas Erskine).— THE CONSTITUTIONAL HISTORY OF ENGLAND since the Accession of George III. 3 vols. Crown 8vo. xZs. MEADE (L. T.).— THE O'DONNELLS OF INCHFAWN. Crown 8vo. 6s. DADDY'S BOY. With Illustrations. Crown 8vo. ■y. 6d. DEB AND THE DUCHESS. Illust. by M. E. Edwards. CT.Zvo.3s.6d. THE BERESFORD PRIZE. lUustratedbyM. E.Edwards. Cr.8vo.SJ. MEATH (The Earl of).— SOCIAL ARROWS: Reprinted Articles on various Social Subjects. ' Crovm 8vo. Jj. PROSPERITY OR PAUPERISM? Physical, Industrial, and Technical Trainii^. Edited by the Earl of Meath. 8vo. sj- MELVTIiIiE (O-. J. Whyte).— Novels by. Crown 8vo. is. each, boards ; Tj. 6d. each, cloth. The Gladiators. The Interpreter. Good tor Nothing. The Queen's Maries. Holmby House. Kate Coventry. Digby Grand. General Bounce. MEMDELSSOHTT,— THE LETTERS OF FELIX MENDELSSOHN. Translated by Lady Wallace, a vols. Crown 8vo. 10s. MEHrVALE (Rev. Chas.).— HISTORY OF THE ROMANS UNDER THE EMPIRE. Cabinet Edition, 8 vols. Crown 8vo. 48s. Popular Edition, 8 vols. Grown 8vo. 31. 6d. each. THE FALL OF THE ROMAN REPUBLIC : a Short History of the, Last Century of the Commonwealth. lamo. 7s. 6d. GENERAL HISTORY OF ROME FROM B.C. 753 TO A.D. 476 Cr. 8vo. 7J. 6d. THE ROMAN TRIUMVIRATES. With Maps. Fcp. 8vo. 2j.' 6d. MIIjES("W".A.).—THECORRESPONDENCEOFWILLIAM AUGUSTUS MILES ON THE FRENCH REVOLUTION, 1789-1817. 2 vols. 8vo. 325. MHiIi (James).— ANALYSIS OF THE PHENOMENA OF THE HUMAN MIND. 2 vols. 8vo. aSf. PUBLISBBP BY MSSSRS. LONGMANS, GSEEN, &= CO. t? MILL (John Stuart).— PRINCIPLES OF POLITICAL ECOIiOMY, AM Library Edition, 2 vols. 8vo. 3ari | People's Edition, i vol. Crown 870.3^. 6;/. -^-7- — A SYSTEM OF LOGIC. . Crown 8vo.'3f. 6d. ON LIBERTY. Crown 8vo. it. ifi. --^ ON REPRESENTATIVE GOVERNMENT. Crown 8vo. 2J. -.UTILITARIANISM, 8vo. SJ. ^— EXAMINATION OF SIR WILLIAM HAMILTON'S PHILO- SOPHY. 8vo. i6j. — NATURE. THE tXTILITY OF RELIGION AND THEISM. Three Essays,' 8v6. tf. , MOLESWOKTH (Mrs.)^MARRYING AND GIVING IN MARRIAGE : a Novel. Fcp. 8vo. zs. dd. •■ SILVERTHORNS. With lUustrations by F. NoelPaton. Cr. 8vo. Jr. THE PALACE IN THE garden: With Illustrations. Cr. 8vo. Jr. THE THIRD MISS ST. QUENTIN. Crown 8vo. (,s. NEIGHBOURS. With Illustrations by M. Ellen Edwards. Cr. 8vo. (,s. THE STORY OF A SPRING MORNING. With Illustrations. Cr.8vo.5j. MOOBE (Edward).— DANTE AND HIS EARLY BIOGRAPHERS. .Crown 8vo. 41. (>d. MUXHALL (Michael G.).— HISTORY OF PRICES SINCE THE YEAR 1850. Crown 8vo. (>s. HAITSEK" (Dr. Fridtjof ).— THE FIRST CROSSING OF GREENLAND. With s Maps, 12 Plates, and 150 Illustrations in the Text. 2 vols. 8vo. 36^. ITAPIEK.— THE LIFE OF SIR JOSEPH NAPIER, BART., EX-LORD CHANCELLOR OF IRELAND. By Alex. Charles Ewald. 8vo. ijf. THE LECTURES, ESSAYS, AND LETTERS OF THE RIGHT HON. SIR JOSEPH NAPIER, BART. 8vo. I2j. td. JSTESBIT (E.).— LEAVES OF LIFE: Verses. Crown Svo. jf. ITEWMAH".— THE LETTERS AND CORRESPONDENCE OF JOHN HENRY NEWMAN during his Life^in the English Church. With a brief ' Autobiographical Memoir. Edited by Anne Mozley. With Portraits, 2 vols. Svo. 305. net. NEWMAN (Cardinal).— Wotks by :— DiBconrsea to mixed Congrafatlona. Cabinet Edition, Crown 8vo. dt. Cheap Edition, 31. fd. Barmgns on Yarlona OccaslonB. Cr. '8vo. 6j. Ibe Idea of a Dniverslty defined and illustrated. Cabinet Edition, Cr. Svo. •js. Cheap Edition, Cr. Svo. 3^. 6i HistbrlcaJ, Sketches. Cabinet Edition, "'3 vols. Crown Svo. ds. each., Cheap ,,- Edition, 3 vols. Cr. Svo. jr. dd. each. The Arlans of th* Fourth Century. Cabinet Edition, Crown Svo. 6;. Cheap Edition, Crown Svo. y. 6d. Select Treatises of St. Athanasius in Controieray nith the Arians. Freely Translated. ^ vols. Crown Svo. DlscusBions and Arguments on'YaTiouG Subjects. Cabinet ^E^dition, Crown 8v6. 6s. Cheap Edition, Crown Svo. 3t. 6d. ' [Contintied. i8 A CATALOGUE OF BOOKS IN GENERAL LITERATURE NEWMAN" (Cardinal)i— "Works by :— (continued). Apololla Pro Vita Sua. Cabinet Ed. , Crown 8vo. 6f. Cheap Ed. y. 6d. Development of Christian Doctrine. Cabinet Edition, Crown 8vo. 6s. Cheap Edition, Cr. 8vo. jr. 6d. Certain Difficulties felt by Anglicans in Catholic Teaching Considered. Cabinet Edition. "Vol. I. Crown 8vo. ;7J. 6d. ; Vol II. Crown 8vo. 51. 6d. Cfieap Edition, 2 vols. Crown 8vo. 3^. dd. each. The Via Media of the Anglican Church, Illustrated in Lectures, &c. Cabinet Edition, 2 vols. Cr. 8vo. 6s. each. Cheap Edition, 2 vols. Crown 8vo. 3J. 6d. each. Essays,CritioaIandHist,oriCal. Cabi- net Edition, 2 vols. Crown 8yo. 12s. Cheap Edition, 2 vols. Cr. 8vo. 7J. Biblical and Ecclesiastical Liraoles. Cabinet Edition, Crown 8vo. 6s. Cheap Edition, Crown 8vo. y. 6d. Present Position of Catholics in Eng- land. Cabinet Edition, Qrown 8vo. 7J. 6d. Cheap Edition, Crown 8vo. 3s. 6d. TreictB. I. Dissertatiunculse. 2. On the Text of the Seven -BpistlfiSJof St. Ignatius. 3. Doctrinal Causes of Arianism. 4. Apollinarianism. 5. ' St. Cyril's Formula:; 6. Ordo de Tempore. 7. Douay Version of Scripture. Crown 8vc>. is. "■' '■ An Essay in Aid of a Gramiuar of Assent. Cabinet Edition, d. THE PHILOSOPHY OF FICTION IN LITERATURE : an Essay. Crown 8vo, 6j. PUBLISHED Bit MESSRS. LONGMANS, GREEN, &■ CO. 23 THREE IN NOIiWAY. . By Two of Thbm. With a Map and 59 Illustratiohs. Crown 8vo. ai. boards ; zs. 6d. cloth. TOYTTBEE (Arnold).— LECTURES ON THE INDUSTRIAL REVO- LUTION OF THE iSth CENTURY IN ENGLAND. 8v6. io.r. , 6i TBEVELYAM" (Sir G. O., Bart.).— THE LIFE AND LETTERS OF LORD MACAULAY. , Popular Edition. Crown 8vo. as. 6ii. I Cabinet Edition, 2 vols. Cr. 8vo. tzs. Student's Edition. Crown 8vo. 6s. | Library Edition, 2 vols, 8vo. 36J. THE EARLY HISTORY OF CHARLES JAMES FOX. Library Edition, 8vo. i8j. Cabinet Edition, Crown 8vo. 6s. TEOLIiOPE (Anthony).— THE WARDEN. Cr. 8v6. is. bds., is. 6(1. cl. BARCHESTER TOWERS. Crown 8vo. is. boards, is. 6d. cloth. VIIiLE (G.).— THE PERPLEXED FARMER: How is he to meet Alien Competition ? Crown 8vo. y. 1 VXEGIIi. — PUBLI VERGILI MARONIS BUCOLICA, GEORGICA,. jENEIS; the Works of Virgil, Latin Text, with English Commentary and Index. By B. H. Kennedy. Crown 8vo. 10^. 6d. THE iENEID OF VIRGIL. Translated into EogUsh Verse. By John Conington. Crown 8vo. 6s. , THE POEMS OF VIRGIL. Translated into English Prose. By John Conington. Crown 8vo. 6s. THE ECLOGUES AND GEORGlCS OF VIRGIL. Translated from the Latin by J. W. Mackail. Printed on Djitch Hand-made Paper. i6mo. ss. "WAKEMAW (H. O.) and HASSAI>Ii (A.).— ESSAYS INTRODUC- TORY TO THE STUDY OF ENGLISH CONSTITUTIONAL HISTORY. Edited by H. O. Wakeman and A. Hassall. Crown Bto. 6s. WALKER (A. Campbell-).- THE CORRECT CARD ; or. How to Play at Whist ; a Whist Catechism. Fcp. 8vo. sj. 6d. "WALPOIiE (Spencer).- HISTORY OF ENGLAND FROM THE CON- CLUSION OF THE GREAT WAR IN 1815 to 1858, Library Edition. 5 vols. 8vo. jf4 iQs. Cabinet Edition. 6 vols. Crown 8vo. bs. eabh. WEIiLIirGTOir.— LIFE OF T^E DUKE OF WELLINGTON. By the Rev. G. R. Gleig. Crown 8vo. js. 6d. WENDT (Ernest Em.il).— PAPERS ON MARITIME LEGISLATION, with a Translation of the German Mercantile Laws relditing to Maritime Com- merce. Royal 8vo. £,1 lis. 6d. VTEYMAN (Stanley J.).— THE HOUSE OF THE WOLF : a Romance. Crown 8vo. 6s. ' WHATEIiT (E. Jane).— LIFE AND CORRESPONDENCE OF ARCH- BISHOP WHATELY. With Portrait. Crown 8vo. los. 6d. WHATELY (Archbishop).— ELEMENTS OF LOGIC. Cr. 8vo. 4s. 6d. ELEMENTS OF RHETORIC. CrowRBvo. 4J. 6d. LESSONS ON REASONING. Fcp. 8vo. is. bd. BACON'S ESSAYS, with Annotations. 8vo. los. 6d. WHiCOCKS (J. C.).— THE SEA FISHERMAN. Comprising the Chief Methods of Hook and Line Fishing in the British and other Seas, and Remarks on Nets, Boats, and Boating. Profusely Illustrated. Crown 8vo. 6s. 24 BOOKS PUBLISHED BY MESSRS. LONGMkNS,fiREEN, ^ CO. WIIiLICH (Charles M.).— POPULAR TABLES for giving Iiirormation for ascertaining the value of Lifehold, Leasehold, and Church Property, the Public Funds, &c. Edited by H. Bence Jones. Crown 8vo. ioj. 6d. ■WILXiOUGHBT (Captain Sir John C.).— EAST AFRICA AND ITS BIGGAME. lllustratedbyG. D.Giles and Mrs. GordonHake. Royal 8vo. su. WITT (Prof.) — "Works by. Translated by Frances Younghusband. THE TROJAN WAR. Crown 8vo. 2j. MYTHS OF HELLAS ; or, Greek Tales. Crown Bvo. 31. 6d. THE WANDERINGS OF ULYSSES. ,Crown Bvo. 3^. 6d. THE RETREAT OF THE TEN THOUSAND ; being the Story of Xenophon's ' Anabasis '. With Illustrations. Crown Bvo. 3^. 6d. "WOIiFr (Henry W.).— RAMBJLES IN THE BLACK FOREST. Crown Bvo. js. 6d. THE WATERING PLACES OF THE VOSGES. With Map. Crown Bvo. ^s. 6d. THE COUNTRY OF THE VOSGES. With a Map. Bvo. xzs. WOOD (Eev. J. O.).— HOMES WITHOUT HANDS ; a Description of the Habitations of Animals. With 140 Illustrations. 8V0. 7^. net. INSECTS AT HOME; a Popular Account of British Insects, their Structure, Habits, and Transformations. With 700 Illustrations. 8vo. js. net. INSECTS ABROAD ; a Popular Account of Foreign Insects, their Structure, Habits, and Transformations. With 600 Illustrations. 8vo. ^s. net. BIBLE ANIMALS ; a Description of every Living Creature mentioned in the Scriptures. With 112 Illustrations. Bvo. js. net. STRANGE DWELLINGS ; abridged from ' Homes without Hands '. With 60 lUustrations. Crown Bvo. 3J. td. OUT OF DOORS ; a Selection of Original Articles on Practical Natural History. With 11 Illustrations. Crown 8vo. 3J. (>d. PETLAND REVISITED. With 33 Illustrations. Crown 8vo. 3J. dd. WORDSWORTH (Bishop Charles).— ANNALS OF MY EARLY LIFE, 1806-1846. Bvo. T.^. WYLIE (J. H.).— HISTORY OF ENGLAND, UNDER HENRY THE FOURTH. Crown Bvo. Vol. I, ioj. 6j?. ; Vol. II. ZELIiEB (Dr. E.).— HISTORY OF ECLECTICISM IN GREEK PHILO- SOPHY. Translated by Sarah F. Alleyne. Crown Bvo. ioi. 6rf. ' '- THE STOICS, EPICUREANS, AND SCEPTICS. Translated by the Rev. O. J. Reichel. Crown Bvo. 15J. SOCRATES AJID THE SOCRATIC SCHOOLS. Translated by the Rev. O. J. Reichel. Crown Svo. loj. td. PLATO AND THE OLDER ACADEMY. Translated by Sarah F. Alleyne and Alfred Goodwin. Crown Bvo. iBj. THE PRE-SOCRATIC SCHOOLS. Translated by Sarah F. AUeyne. 2 vols. Crown Svo. 30J. OUTLINES OF THE HISTORY OF GREEK PHILOSOPHY. Translated by Sarah F. Alleyne and Evelyn Abbott. Crown Bvo. \os. 6d. 50,0CX>— 2/92. ABERDEEN UNIVERSITY PRESS.