^^-^c 
 
 S""^ 
 
 "Una. 
 
 e^^:^^: ^ 
 
 mi:m^ 
 
 •^:j^m 
 
 ^•vCv"^ 
 
 .4 
 
 
 
 '': ^ 
 
 ^' 
 
LIBRARY OF CONGRESS. 1 
 
 ^i 
 
 €l,ap. T/^53 i 
 
 s^„,, ^^ 
 
 I UNITED STATES OF AMEEICA. I 
 
t 
 
 HORA PHILOLOGICA, 
 
 OR, 
 
 CONJECTURES ON THE STRUCTURE 
 
 OF THE 
 
 GREEK LANGUAGE. 
 
 BY WILLIAM SEWELL, M. A. 
 
 FELLOW OF EXETER COLLEGE, OXFORD. 
 
 OXFORD, 
 D. A. TALBOYS 
 
 1830. 
 
^h 
 
 
 ^'K 
 
 TALBOYS AND BrxOWNE, PRINTERS, OXFORD. 
 
TO THE 
 
 SOCIETY OF EXETER COLLEGE, 
 
 THIS LITTLE ESSAY 
 
 MOST GRATEFULLY AND AFFECTIONATELY INSCRIBED 
 
 BY THE AUTHOR. 
 
NOTE TO THE READER. 
 
 In the only place where a writer is allowed 
 to speak of himself, I may, perhaps, be per- 
 mitted to apologize for any errors in the fol- 
 lowing little Essay, first, by candidly plead- 
 ing very great ignorance of the critical labours 
 of others ; and secondly, by explaining that 
 the observations themselves were made solely 
 for my own amusement, during a late tour on 
 the continent, when I could obtain no access 
 to books ; and possessed no other means of as- 
 certaining their accuracy, but conversation with 
 a most valued and gifted friend *, to whose ta- 
 lents and researches on similar subjects I may 
 attribute either directly, or indirectly, anything 
 not altogether valueless in the conjectures 
 themselves. 
 
 As I am not likely to possess much leisure 
 for prosecuting philological inquiries, and have 
 found even the present sketch, with all its ma- 
 nifold imperfections, useful on more than one 
 occasion in encouraging a fondness for critical 
 studies, I have not scrupled, though with great 
 diffidence of its value, to offer it to any student 
 who may be inclined to exercise his own mind 
 in refuting or confirming the hypotheses of 
 others. 
 
 * William Crichton, esq. of Merton college, Oxford. 
 
HORA PIIILOLOGICA 
 
 USE OF LANGUAGE. 
 
 Among the many principles of action, which 
 Nature has implanted in the Human mind, 
 there are two which seem principally to have 
 led us to the formation and employment of 
 Language ; our animal wants, and the desire of 
 Sympathy. If we could supply the former 
 without the assistance of society, or gratify 
 the latter without communicating our ideas, in 
 all probability language would have never ex- 
 isted. In our present constitution, however, 
 we find it perpetually necessary to act upon 
 the minds of the beings which surround us, in 
 order to bring them into those states in which 
 they will be inclined to satisfy our wants, or 
 harmonize with our own feelings. And man, 
 thus surrounded by animated creatures, is not 
 unlike a musician playing on an enormous organ 
 filled with millions of chords and pipes, and 
 moving on the most delicate springs, which run 
 into and act upon each other in complications 
 impossible to follow ; and produce either har- 
 mony or discord in proportion to the skill of 
 the performer. 
 
 B 
 
2 USE OF LANGUAGE. 
 
 Now as every impression on the mind must 
 be made through one of the senses, we must 
 employ as the medium of our communication 
 with it, either the touch, the taste, the smell, 
 the eye, or the ear. We employ the first when 
 we whip a child to keep him from mischief; 
 the second, when we bribe him with sugaf- 
 plums ; the third, when we apply scents to 
 the nose of a fainting person; the fourth, in 
 exhibiting a picture ; and the fifth, in the 
 composition of music. And in all these cases 
 the mind is immediately brought into the state 
 which we wish to produce. 
 
 But as our object is to possess an absolute 
 control over its habits, and the power of bring- 
 ing it, as we wish, into every possible state, of 
 which it is susceptible ; it is evident that both 
 the organs which we affect, and the instru- 
 ments by which we act on them, must severally 
 possess the two following qualities : 
 
 I. They must be as much as possible within 
 our own command. 
 
 II. They should be capable, the one of con- 
 veying, and the other of receiving, as great 
 a variety of distinct impressions, as there are 
 distinct ideas to be produced in the mind. 
 
 To possess a perfect command of musical 
 combinations we must have an instrument pro- 
 vided with the requisite number of keys, and 
 those keys completely within our reach. 
 
USE OF LANGUAGE. 3 
 
 With respect to the first qualification, it is 
 obvious that the ear is the organ to which we 
 have most readily access : we can reach it 
 with greater rapidity; from a greater distance 
 than any other ; under circumstances when no 
 'other can be approached; and with the least 
 probability of finding its avenues blocked up. 
 The slightest partition will prevent our acting 
 upon the touch, the taste, the smell, or even 
 the eye ; and any one of these senses we can 
 easily prevent from receiving impressions, when 
 disinclined to admit them. The ear is however 
 always open, and difficult to be closed. And 
 the instrument, by which we are enabled to act 
 on it, is likewise more in our power than any 
 other of our organs ; since it is the hardest of 
 all to confine, and nearly the last of which we 
 are deprived by dissolution. 
 
 With regard to the second qualification, sup- 
 posing that we were able to produce none but 
 immediate impressions on the mind ; i. e. none 
 through the ear but sounds, none through the. 
 eye but colour and figure, and none through 
 the touch but heat and cold, or hardness 
 and softness, it is evident that the fittest me- 
 dium for our mutual communication would be 
 the sense which took the widest range, and 
 admitted the greatest variety of ideas. This 
 would of course be the eye ; and in such a 
 language actual substances, or pictures, would 
 
 B 2 
 
4 USE OF LANGUAGE. 
 
 constitute our words ; as travellers converse 
 with savages, and the Mexicans transmitted 
 their news. But such a language would be in 
 the first place an extremely clumsy machine. 
 Portable mountains, and rivers, and trees, are 
 not procurable at a moment's notice, when we 
 wish to set them before the eyes of our audi- 
 ence ; and even pen, ink, and paper, to sketch 
 them, are frequently out of our reach. 2. It 
 would be very deficient, since all the art of 
 man could not convey through the eye the 
 ideas of sweetness, bitterness, grave sounds, 
 coldness or warmth, or, in short, any other 
 ideas than those which we perceive through 
 the eye. And thus the other four senses would 
 be left without any representative. 3. It would 
 fail in that essential point, the power of con- 
 veying abstract ideas, since to the eye colour 
 could never be exhibited without extension, 
 nor extension without figure ; nor a quality of 
 the mind without an action of the body; nor 
 motion without a number of accidental circum- 
 stances ; so that the process of distinguishing 
 the idea we wished to call up, amidst this 
 group of extraneous attributes, would be one 
 of considerable time, and no less difficulty''. 
 Fortunately from this awkward resource Na- 
 ture herself has rescued us by the law of asso- 
 
 * See Appendix (A). 
 
USE OF LANGUAGE. 5 
 
 elation ; that simple law, by which when the 
 mind has passed several times consecutively 
 from one state to another, from A to B, it 
 acquires a tendency to repeat the same order 
 of ideas whenever the first link occurs ; and 
 to slide into A whenever it is brought into B. 
 By this law then we are enabled to select 
 one single sense, in order to convey a number 
 of impressions to the mind ; each of which im- 
 pressions, having previously been associated with 
 some other idea, will call it up again in future, 
 and stand to it in the relation of a sign, or in- 
 dicator, as a buoy to an anchor, a rope to a 
 bucket in a well, or a bank note to the value 
 which it represents. And taking this last kind 
 of sign as an illustration, we shall be able to 
 ascertain by it what are the requisite qualifica- 
 tions of signs in general. One great advantage 
 then of paper money as a medium of exchange 
 is its infinite divisibility, by which it is enabled : 
 1 . to represent every possible denomination of 
 value; 2. to represent them distinctly and 
 precisely; and 3. to effect this by a very sim- 
 ple and easy process; viz. the repetition and 
 combination of a very few primary elements. 
 Another great advantage is the rapidity with 
 which it can be circulated; and the certainty 
 and simplicity with which our calculations may 
 be carried on in it, without a constant reference 
 to positive value incorporated in articles of 
 
6 USE OF LANGUAGE. 
 
 commerce. Another advantage is its perfect 
 intrinsic valuelessness ; by which, if properly 
 managed, it would be preserved from all fluc- 
 tuations ; and prevent any confusion from 
 arising between the sign of value and the thing- 
 signified. 
 
 Now precisely analogous qualifications are 
 requisite in those ideas or impressions, which 
 we employ as signs to the mind, or representa- 
 tives of other ideas. 
 
 I. They should be sufficiently numerous to 
 supply a different sign for every different idea 
 or group of ideas. 
 
 II. Each of these signs should be distinct 
 from the other, in order to call up distinct ideas. 
 
 III. As their variety must be very great 
 from the nature of their office, and yet it is 
 difficult for the memory to retain any consider- 
 able number, they must be compounded of a 
 few simple elements, whose signification may 
 easily be embraced and retained in the mind. 
 
 IV. They should be capable of being em- 
 ployed with quickness and facility, to satisfy 
 the impatience of a substance which has to 
 express the most rapid movements by a slow 
 vehicle. 
 
 V. They should be as abstract as possible, 
 to prevent any confusion, and to enable us to 
 employ them with precision and distinctness. 
 
 It is evident that in comparing the means 
 
FORMATION OF LANGUAGE. 7 
 
 which the eye and the ear respectively present 
 of conveying to the mind a series of signs thus 
 constituted, the ear possesses a great superiority 
 over the eye in all cases where we can reach 
 by sounds the object which we wish to affect. 
 Where it is removed to a distance we address 
 ourselves to the eye ; and in both cases we 
 endeavour to frame signs endowed with all the 
 requisite excellences. Hence in speaking we 
 are glad to employ a few primitive sounds, 
 which are very rapidly pronounced, run easily 
 into each other, present but one single idea to 
 the mind, and represent an infinite variety of 
 ideas by combinations of the same simple 
 principles. And in writing, the same. — Each 
 letter being as simple as possible, as quickly 
 written, as unmeaning in itself, and as rich in 
 the multiplicity of its uses, as it is numerous in 
 the possibilities of its combinations. 
 
 Still, notwithstanding the judgment with 
 which sounds have been selected as the me- 
 dium of our mutual communications in society, 
 it remains to be inquired, how we were led to 
 select and to frame them. It is quite obvious 
 that the first man who employed language had 
 never contemplated the subject in this light, or 
 weighed all the relative merits and demerits of 
 the different senses. It is equally difficult to 
 conceive how any two or more persons could 
 have existed without some means of mutual 
 
8 FORMATION OF LANGUAGE. 
 
 communication ; or how, without some means 
 already established, they could arbitrarily 
 agree to attribute a certain number of ideas 
 respectively to certain insignificant signs. It 
 is also a maxim of sound philosophy, that 
 where a phenomenon may be explained by 
 facts which are clear and ascertained, we are 
 not to wander for solutions into the mists of 
 hypothesis and conjecture : and lastly, there is 
 a strong probability that Nature, who has so 
 elaborately provided for all the other wants of 
 our helplessness, has not left us unassisted in 
 this most essential particular ; but having made 
 language necessary to our happiness, led us 
 also instinctively to its formation. And if we 
 can once discover laws of mind adequate to 
 accomplish this purpose, we may consider the 
 problem as solved, and neither have recourse 
 to revelation where revelation is not necessary, 
 nor to the arbitrary caprice of man where such 
 a principle could not possibly operate. 
 
 We find, then, that Nature has placed in our 
 throats a very curious and complicated ma- 
 chine, by which, with the assistance of the 
 organs of the mouth, we are enabled to pro- 
 duce a number of sounds, and to modify, ar- 
 range, and combine them as we wish : we feel 
 also that there is a strong sympathy, or, per- 
 haps to speak more accurately, a close connec- 
 tion between this instrument and other nerves 
 
FORMATION OF LANGUAGE. 9 
 
 and muscles communicating with the brain ; 
 so that an affection of one of these nerves is 
 immediately followed by an affection of the 
 voice ; and this not indiscriminately, but with 
 all the distinctions of feeling accurately mark- 
 ed. Joy never is expressed in slow and low 
 tones, or grief in rapid and lively airs. Parti- 
 cular notes are appropriated to particular pas- 
 sions. The very rhythm of sounds indicates 
 the state of the mind; and their mere inflexions 
 and cadences will not only express, but com- 
 municate the several emotions which produced 
 them. The action in short is reciprocal. Sor- 
 row in the mind of one man produces a flood 
 of tears, and that flood of tears instinctively 
 produces sorrow in the mind of another. Tri- 
 umph is expressed by a loud cry of exultation; 
 and this same sound produces in other minds 
 the same state from which it originated itself. 
 Beyond this fact we cannot perhaps at present 
 advance, but we here obtain one element of 
 language instinctively taught us without our 
 choice ; without even our consciousness of the 
 mode by which to employ it. We have certain 
 simple sounds, and modifications of sounds, to 
 express and to communicate certain passions 
 of the mind. 
 
 II. The love of imitation, which is perhaps 
 resolvable into three principles : the force of 
 sympathy, the tendency of the mind to recur 
 
10 FORMATION OF LANGUAGE. 
 
 into a previously existing state, and in some 
 cases the pleasure of exercising power, sup- 
 plies us at once with another class of words, 
 which represent sounds by sounds. 
 
 Onomatopoeias abound in all tongues, and 
 are formed every day by the most vulgar and 
 ignorant minds. 
 
 IIL As some sounds are pleasing and others 
 displeasing, it will be very natural, by the law 
 of association, to represent all agreeable ob- 
 jects by the first class of sounds, and dis- 
 agreeable objects by the second ; and the 
 influence of this analogy is perceptible in all 
 languages. 
 
 IV. Not only will individual ideas be repre- 
 sented in this manner, but the mode of succes- 
 sion in which they occur, the quickness, or 
 slowness, or abruptness of their transition, will 
 be likewise expressed by quick, or slow, or 
 abrupt sounds. The rhythm in which the 
 ideas are conveyed to the mind, will exactly 
 tally with the rhythm of the sounds by which 
 they are indicated. It is quite as unnatural to 
 dance out of time as to be sorrowful in tro- 
 chaics or joyful in spondees. 
 
 V. There is a great sympathy between the 
 different senses; and the man who compared 
 the colour of scarlet to the note of a trumpet, 
 from finding that the impression on his mind 
 was in some degree the same, would naturally 
 
FORMATION OF LAxVGUAGE. II 
 
 express the one by the other. Every one has 
 heard of tastes being compared to odours, and 
 feelings to sounds. Thus we speak of harsh 
 outlines and harsh tones, of clear light and 
 clear notes, of softness as applied to the touch, 
 the ear, and the eye. And the universal mix- 
 ture of metaphor w^hich pervades all languages 
 w^hen speaking of the senses, is a sufficient 
 illustration of the fact. 
 
 VL Accidental and natural associations will 
 frequently have connected sounds with ideas 
 which naturally are totally distinct from them^. 
 
 VII. The slightest analogies between objects 
 will be seized, and give rise to their classifica- 
 tion under one common sound. 
 
 VIII. As we employ in shorthand writing, 
 part of a letter to indicate a whole word, so in 
 the wish to express our thoughts with rapidity 
 equal to their succession, we shall employ a 
 single sound to express a whole group of ideas 
 which we have observed in connection with it : 
 the rustling of a tree to represent a tree, a 
 sound like a roar to indicate the sea. 
 
 Such are some of the chief principles which 
 would guide man in the adoption of particular 
 sounds to represent ideas and objects. None 
 of them arbitrary, none depending on our own 
 choice. None even, in which we are ac- 
 
 *> 3ee Appendix (B). 
 
12 DIVISION OF WORDS. 
 
 quainted with the mode of operation. The 
 meanest rustic acts upon them every day in 
 the communication of his feelings : and if an- 
 other world were peopled from a single pair 
 constituted and organized as man, its language, 
 with very few exceptions arising from acci- 
 dental associations, would be the same as the 
 one which we employ. 
 
 Besides, however, the use and advantage of 
 language and the means by which we arrived 
 at it, there is another inquiry concerning the 
 nature of those combinations of sound, whether 
 expressed by the voice or represented by writ- 
 ten lines, which we denominate words. And 
 difficulties have been thrown in the way of this 
 inquiry by the very nature of language. In 
 learning to speak we do not acquire each word 
 separately, attaching to each its peculiar signi- 
 fication. But we run off whole sentences by 
 rote, in which the only terms invested with 
 distinct ideas are objects of sight, and of the 
 other senses, and all the abstract terms, with 
 the intermediate links, and little particles 
 which tie them together, convey nothing but 
 a vague and scarcely perceptible notion of 
 something very undefined, which we do not 
 take the trouble to examine. The sounds run 
 so easily and quickly into each other both from 
 habit and their own natures, that we never 
 stop to break their links, and examine each of 
 
DIVISION OF WORDS. 15 
 
 them apart from the rest. And hence three 
 errors have been fostered in many grammatical 
 speculations. 
 
 I. The opinion that one word influences or 
 governs another, because they are generally 
 connected together, just as the conjunction of 
 antecedents and consequents in the physical 
 world has given rise to the vulgar belief in 
 some mysterious energy communicating be- 
 tween them. And the real source of both 
 these errors may perhaps be traced to the 
 rapid action of the mind, which actually blends 
 and runs together states which have frequently 
 appeared in succession. 
 
 II. The assertion that some words have no 
 signification except when combined with others. 
 
 III. The belief that son^e have no original 
 signification whatever, because none has been 
 attached to them by us. 
 
 Whereas if we look to the earliest languages 
 and those least perfect in the present day, we 
 find them made up of words entirely distinct 
 from each other, each with a separate meaning, 
 and unconnected by any particles. The more 
 also we analyze by etymology our own more 
 complicated tongues, the more reason we find 
 to break them up into their separate elements. 
 And in fact, since every word was first em- 
 ployed as a sign of some state of mind, and 
 those states of mind follow one another in re- 
 
U DIVISION OF WORDS. 
 
 gular but distinct succession, their signs must 
 be equally distinct in their original use and ap- 
 plication. And to say that a sign was em- 
 ployed without anything for it to signify, that 
 words were framed to represent things which 
 did not exist, which is to say that there are 
 words without any meaning properly attached 
 to them, is as great an absurdity as to conceive 
 the institution of a paper currency without 
 value attached to it, or of a telegraphic com- 
 munication, where neither party understood 
 the movements. 
 
 Three principles then must necessarily be 
 laid down. 
 
 I. That every sound or separate combina- 
 tion of sounds was originally invested with a 
 distinct, appropriate, and self-existing meaning. 
 
 II. Upon the same grounds, that the slightest 
 variety of inflexion, where it cannot be traced 
 to a confused pronunciation, indicates a differ- 
 ence of meaning. 
 
 III. That, every sound having originally but 
 one meaning, wherever it appears to compre- 
 hend many, some analogy must have existed 
 between them. 
 
 In dividing, however, words into their seve- 
 ral classes according to the nature of the ideas 
 which they represent, much confusion has 
 gei3Lerally been produced, in some instances 
 by rejecting altogether the metaphysics of 
 
DIVISION OF WORDS. 15 
 
 language, or to avoid a much-abused and 
 ill-understood term : an investigation of the 
 human mind as connected with the employ- 
 ment of words; and in other cases by confining 
 the inquiry to the material and external world ; 
 and bringing a cloud over the subject as thick as 
 ^Egyptian darkness, by playing with the words 
 essence, substance, attribute, accident, pro- 
 perty, and all those logical terms, to which no 
 definite meaning has ever been attached ; and 
 which are employed about things very far re- 
 moved from our cognizance ; and whose very 
 existence is disputable. Instead therefore of 
 classifying words according to their accidental 
 form, or local position, or the imaginary attri- 
 butes of matter ; and of consequently throwing 
 into one confused and neglected heap, a vast 
 number denominated particles, which from our 
 own ignorance are unsusceptible of any other 
 arrangement, it will be advisable to reduce 
 them under as few heads as possible, accord- 
 ing to the different states of mind, which they 
 serve to represent. 
 
 And to do this we may propose the follow- 
 ing axioms : 
 
 I^ That every simple idea, perception, sen- 
 sation, or notion, or whatever other term we 
 may choose to employ, is nothing but the mind 
 
 c See Appendix (C). 
 
16 DIVISION OF WORDS. 
 
 itself in a particular state; just as a piece of 
 wax moulded now into a cube, now into a tri- 
 angle, and now into a globe, is still the same 
 wax under a different shape. Of course the 
 analogy does not extend to imply the mate- 
 rialism of the mind. 
 
 II. That the mind can be but in one state, 
 or be conscious of one simple idea at a time. 
 
 III. That the succession of these states is 
 regulated first by the succession of external 
 impressions, and next by the law of association. 
 
 IV. That the mind has the power of cutting 
 the chain wherever it chooses ; of detaching 
 from it one link, or two, or three, or as many 
 as it likes ; into which power we may resolve 
 the faculty of abstraction. 
 
 V. That it frequently employs a single sound, 
 or one state of mind to indicate and stand for 
 any group which it has thus detached. 
 
 VI. That every group thus detached becomes 
 in language a substantive, and every member 
 when considered as a part of it, becomes an 
 adjective, as a tall fair man, a green spreading 
 beautiful tree, a clear deep silent river ; where 
 each of the groups constitutes but one sub- 
 stantive composed of several ideas, which the 
 mind runs through connectedly, till it comes to 
 a termination ; and then breaks them off from 
 the ideas which are ready to follow — a princi- 
 ple which perhaps may be illustrated by the 
 
DIVISION OF ^^ORDS. 17 
 
 fact, that when we convert an adjective into a 
 substantive, or a concrete noun into an abstract, 
 we generally denote the change of signification 
 by some alteration at the end of the word, not 
 at the beginning. 
 
 VII. That whenever anyone adjective mem- 
 ber of a group is peculiarly striking and impres- 
 sive, it fixes the mind upon its contemplation, 
 and thus becomes a substantive still connected 
 with the former elements of the compound 
 series. Thus, we say a man of justice, instead 
 of a just man ; a tree of verdure, instead of a 
 green tree; a vision of beauty, instead of a beau- 
 tiful vision. — And this conversion of an adjec- 
 tive idea into a substantive is instinctive ; and , 
 takes place upon the same principle as one 
 part of an object, or one quality amidst a num- 
 ber, when contemplated through the senses, 
 arrests the attention of the mind, and makes us 
 pause and dwell on it without passing on to 
 any other. And it is thus that the use of ab- 
 stract nouns instead of concrete, is not only 
 indicative, but productive of greater strength 
 and energy of thought. 
 
 VIII. That visible objects will supply us on 
 all occasions with the most easy and natural 
 source for analogous descriptions. The act of 
 defending will be indicated by the position of 
 the person before or over the object protected. 
 — equality will be denoted by the supposed 
 
18 DIVISION OF WORDS. 
 
 situation of the parties face to face — inferiority 
 by the relation of lowness ; a state of the mind 
 by a state of the body, as terror by shuddering, 
 fear by paleness, cowardice by the act of run- 
 ning away — so also the abstract feelings and 
 operations of the mind will be expressed by the 
 visible and tangible external objects which tend 
 to produce similar sensations. The state of the 
 mind, when uncertain how to act, will be spoken 
 of as wavering, suspended, balanced — the terms 
 inclined, urged on, propelled, conception, ima- 
 gination, in fact the whole vocabulary, which we 
 employ when speaking of mental operations, is 
 drawn from matter and from sight ; and incalcu- 
 lable mischiefs have arisen from this source both 
 to morals and philosophy. How we are led to 
 the employment of these metaphors is obvious. 
 - —The range of the eye, and the multiplicity of 
 the ideas conveyed by it, readily connect visible 
 objects even with the most abstract notions. 
 By far the greater part of these notions are ex- 
 cited solely through the medium of sight. The 
 states of mind produced by the physical opera- 
 tions of our senses, are in very many instances 
 similar, if not the same, with our moral and 
 intellectual sensations. — To which we may add 
 the clearness and definiteness of such ideas 
 themselves : and the facility of removing any 
 obscurity by the assistance of gesture and 
 action. 
 
DIVISION OF WORDS. 19 
 
 IX. That the mode in which words have 
 passed in every language from one simple idea 
 to the representation of a number, and back 
 again to the simple or abstiact idea, is the fol- 
 lowing : — The word in its first employment ex- 
 pressed one single quality, which was taken to 
 denote an aggregate of other qualities combined 
 in an individual object. But by the law of asso- 
 ciation, whenever a similar object occurred, the 
 word would recur likewise. But the similarity 
 not amounting to identity, the feature of differ- 
 ence would instinctively be struck off from the 
 group which the word denoted. And thus with 
 every repetition of a similar object some new 
 feature would be detracted, till, if the word 
 was in common use, it would come to sig- 
 nify nothing, but those few and remote resem- 
 blances in which every single instance con- 
 curred. Hence it is that the commonest terms 
 are those to which it is most difficult to affix a 
 precise and definite meaning, since such a 
 definition can only be attained by a very ac- 
 curate and extensive comparison. While at 
 the same time it is in these cases that we are 
 least likely to demand a definition, since the 
 frequency of the repetition, and the consequent 
 absence of all hesitation, deludes us into a 
 notion, that the rapidity of sound is the result 
 of certain knowledge. Any one may try the 
 experiment by asking himself the meaning of 
 
 c 2 
 
20 DIVISION OF WORDS. 
 
 the common particles and conjunctions, which 
 enter into every sentence we utter; and the 
 lesson will be still more impressive, if we put 
 before ourselves such words as truth, time, 
 motion, action, place, freedom, and the like, 
 which every one perpetually uses, and which 
 very few profess to understand. 
 
 If these axioms are true, it follows that every 
 word originally expressed some one state of 
 mind, which if considered with reference to 
 the object which produced it, we call a quality. 
 That it represented either a simple idea ab- 
 stractedly, or a group collectively ; and the 
 number included in that group not being ac- 
 curately defined, but being sometimes greater 
 and sometimes less, according to the accidental 
 differences of persons and objects, from hence 
 results the confusion of language, and the many 
 errors which it has occasioned and sanctioned. 
 
 Again, every word will be an idea, or group 
 of ideas, detached and separated from, or consi- 
 dered in connection with some other, and thus 
 we arrive at one primary division of words into 
 substantives and adjectives. And as all rea- 
 soning when divested of its mystery is nothing 
 but a succession of ideas, and we have no con- 
 sciousness except when an idea is in the mind, 
 or rather the mind in an idea, and these ideas, 
 as before explained, must be either detached 
 from each other, or run and be dovetailed into 
 
DIVISION OF WORDS. ^1 
 
 each other, every word, under whatever shape, 
 must ultimately be resolvable into a substan- 
 tive or adjective. 
 
 But as in the chain of our ideas many links 
 repeatedly recur, which for various reasons it 
 is necessary to distinguish ; since on one occa- 
 sion a predicate may be applicable to them 
 which is not so at another, and as these similar 
 ideas or groups of ideas, are either individuals 
 of the same kind, or the same individual at dif- 
 ferent moments, we are obliged to set a mark 
 upon the one which we wish to point out, by 
 annexing to it some other idea which is found 
 in juxtaposition with this one only. If a chain 
 was put into my hand, of which the links were 
 represented by the letters of the alphabet, and 
 the letter D recurred ten times; when I desired 
 to point out any one letter D in particular, I 
 could only do it by subjoining another letter 
 close to which it was found ; as the D which 
 follows E, the D which precedes P, or the D 
 which is two letters off from K. The letter 
 itself being the same, and distinction being only 
 attainable by creating a difference, that differ- 
 ence must be sought in something externaL 
 And just so in the case of our ideas. 
 
 Again, in forming our complex groups of ideas 
 it is frequently necessary to combine them of 
 two or more substantives — hung together upon 
 one main and central substantive. As a tree of 
 
22 COMPLEX SUBSTANTIVES. 
 
 great height on the hill near the river — and the 
 Greek language possesses a peculiar facility of 
 throwing these accessory ideas into the form of 
 adjectives. This mode of description is more 
 precise and less liable to mistake, than when 
 all the adjuncts are broken up into separate 
 substances ; but it is less energetic, and either 
 fatigues the mind by keeping it in suspense, or 
 disappoints it by dragging a long train of appa- 
 rently useless epithets after the principal sub- 
 ject. It may also be observed, that the idea 
 attached to the chief noun may be either one 
 contained in it, or distinct from it — a tree of 
 great height, or a tree on the hill, which may 
 be expressed in Greek either by the combina- 
 tion of a substantive and an adjective, or by a 
 group of two substantives. So also in English, 
 we may substitute a very high tree for a 
 tree of great height, and in old language, a 
 hill tree for a tree on the hill. But the sub- 
 stantive height, expresses a quality supposed to 
 exist in the tree, while the substantive hill, is 
 an idea totally distinct from it. Hence it is 
 that substantives are mostly employed as ap- 
 pendages in descriptions of the second kind. 
 Now whenever, either for the purpose of 
 distinction or description, we annex a second 
 substantive to a former one, we find, that in 
 Greek the second is marked out by a slight 
 inflexion of the primitive form of the word — by 
 
GREEK CASES. 23 
 
 what are denominated cases. In English, as 
 in many other tongues of northern origin, these 
 cases do not exist ; and the dependence of one 
 noun on another is denoted by a little prefix, or 
 preposition. These prefixes exist in Greek 
 likewise, and in that tongue are much more 
 numerous and complicated than they are in 
 ours. But in both, from the frequency of their 
 appearance, and the consequent abstractedness 
 of the ideas associated with them, it is very diffi- 
 cult to ascertain their precise and original sig- 
 nification. Some approach may however be 
 made to the peculiar meaning both of the 
 Greek cases and Greek prepositions, if the fol- 
 lowing principles are admissible. 
 
 I. It is very improbable that any language 
 should contain two diff'erent forms for express- 
 ing precisely the same thing. Synonyms are 
 rarely identical, and when they are so they 
 may generally be traced back to two distinct 
 tongues. It is, therefore, not likely that the 
 Greek cases express those relations which are 
 denoted by the prepositions. 
 
 II. If the case comprised any idea distinct 
 from and superadded to the full meaning of the 
 nominative, as in all other instances of com- 
 pound nouns, such an addition would be 
 marked by the annexation of a separate ele- 
 ment to the original word. To express a com- 
 plex idea we should use a compound term. 
 
24 GREEK CASES. 
 
 But the cases are distinguished from the nomi- 
 native, solely by a slight inflexion of the termi- 
 nating sound. It is, therefore, natural to think 
 that they denote nothing but the same idea 
 with the nominative, slightly varied in the 
 mode of contemplation. 
 
 III. As the verb and the noun express the 
 same radical ideas, with this distinction only, 
 that the verb comprises also the place of the 
 idea in the order of time, and the person of the 
 subject, it is probable, that a close analogy 
 exists between their respective inflexions ; and 
 that if we find the idea in the verb placed in 
 three different lights, each discriminated by a 
 peculiar inflexion ; and if these three lights 
 are clearly ascertained there, we may transfer 
 the same notions to the inflexions of the noun, 
 and try without presumption the validity of the 
 hypothesis. And this becomes still more pro- 
 bable when we find that the languages which 
 possess cases, possess also inflexions for these 
 tenses, and that where these are wanting the 
 others are wanting also. 
 
 IV. As our whole mental existence is a series 
 of successive perceptions or states of mind, fol- 
 lowing each other as the links in a chain or the 
 letters of an alphabet ; and as there can be no 
 immediate connection except between pairs, as 
 A and B, C and D, let us place any substantive 
 idea in all the several positions which it can 
 
GREEK CASES. 25 
 
 occupy, in such a series, and examine whether 
 both in nature and in number they correspond 
 with the inflexions of the Greek noun. 
 
 And first it may be observed, that in all such 
 successions of pairs, the only idea which is 
 presented to the mind in different shapes and 
 lights is the second and final. The first link in 
 the couple universally slides past under the 
 same appearance. But the check which is 
 given to the thought by coming to a close, 
 throws it back upon the ultimate point, just as 
 boys are found to dwell upon the last word in 
 a sentence, when the continuation is lost to the 
 memory ; and we all of us naturally linger at 
 the termination of any employment. In some 
 instances, indeed, a single idea is presented to 
 the mind in these three different forms, as in 
 exclamations of surprise, admiration, or suffer- 
 ing, and the cases vary accordingly. If this 
 principle be true, we should expect a priori 
 that the first idea in all those pairs which, as 
 substantives, are linked together, should always 
 be expressed by one uniform inflexion. And 
 that the second only should be subject to 
 modification. And this second we should find 
 would appear under two different shapes ; that 
 is, it may either occur but once, or it may 
 vibrate, as it were, upon itself, and be repeated 
 several times. In the former case also it may 
 be susceptible of two positions ; for it may 
 
26 GREEK CASES. 
 
 either come full and distinct and with a sort of 
 totality and completeness upon the preceding 
 idea, or it may be detached from it by an in- 
 terval, which may serve to suspend the mind, 
 and present to it but a vague, indistinct, in- 
 ceptive, and anticipated perception. 
 
 To repeat the statement. If all our substan- 
 tive perceptions occurred singly, the noun 
 would perhaps admit of no inflexions of case. 
 But as frequently, for the purpose of descrip- 
 tion and distinction, they are placed together in 
 pairs, the first of the two will occur in one uni- 
 form shape, which is termed the nominative 
 case ; the second will admit of three inflexions, 
 accordingly as the idea occurs but once, or re- 
 peatedly without extraneous interruption ; and 
 if but once, accordingly as it is presented fully 
 and completely to the mind, or is kept at a 
 distance and suspended by any imaginary in- 
 terval. Precisely in the same manner as we 
 find the being, the action, or the passion repre- 
 sented by the Greek verb, is placed in three 
 different lights, as perfect, imperfect, or incep- 
 tive. Perfect, when it occurs as a whole and 
 completed unity. Imperfect, when it is con- 
 ceived in duration, that duration implying an 
 uninterrupted repetition of a primary idea. 
 Inceptive, when it is looked forward to as 
 through a vista, as something about to be done, 
 or about to exist, and not yet fully brought 
 
GREEK CASES. 27 
 
 before us. If it is concluded that this notion 
 of substantive ideas or groups of ideas being 
 connected only in pairs cannot be true, because 
 frequently a whole cluster are massed together 
 in the same description, as in the sentence, 
 *' Cicero was killed in his litter near the sea by 
 the officers of Antony ;" it is to be remem- 
 bered, that sometimes the primary idea or 
 nominative case is to be repeated to two or 
 more sequences, and that where this is not the 
 case, the sequences are themselves connected 
 in couples. To acknowledge, however, that 
 the mind cannot be sensible of more than one 
 idea at a time, implies that their succession 
 must be carried on in the manner before de- 
 scribed. 
 
 Assuming, then, as an hypothesis, the above 
 explanation of the Greek cases, and remember- 
 ing, that whenever we endeavour to ascertain 
 the meaning and intentions of ages long since 
 past away, where no positive explanation has 
 been transmitted to our hands, an hypothesis 
 is all that we can attain ; we might proceed to 
 try its application and validity in deciphering 
 the precise meaning of the combinations in 
 which these cases occur. To trace it out 
 through all its ramifications is by no means the 
 object of the present sketch. The study of 
 dead languages is principally valuable from the 
 exercise which it affords to the mind ; and 
 
28 GREEK CASES. 
 
 every conjecture should be left to be sanc- 
 tioned or refuted by the observation of the in- 
 dividual inquirer. A few hints, however, may 
 be thrown out. Let the genitive case indicate 
 the second of two substantives conceived in its 
 unity and totality. The dative, the same sub- 
 stantive as removed by an interval from the 
 former, and presented as incipient and pro- 
 spective. And the accusative the same like- 
 wise, when contemplated as recurring upon 
 itself. A single note in music may represent 
 the first case ; one which, from previous asso- 
 ciation, we are ready to anticipate and slide 
 into, the second ; and a long-protracted, unin- 
 terrupted swell may illustrate the third. 
 
 The genitive then will take the widest range, 
 and admit of the greatest variety of instances. 
 It will express the second of any two substan- 
 tive objects considered in the relation of rest ; 
 the point from which motion proceeds, since 
 amotion implies the contemplation of such a 
 point in its extremity, and consequently as a 
 whole ; the point to which motion proceeds, 
 when the motion is made to the extreme. And 
 all the varieties of substances from which mo- 
 tion can be conceived to arise, will also be in- 
 cluded in the same case. 
 
 The dative again will imply an object merely 
 approximating to a former — where an interval 
 exists between them ; the circle within which 
 
GREEK CASES. 29 
 
 a substance is contained, since there such an 
 interval necessarily exists ; the point towards 
 which motion is making, when such a point is 
 indefinite and uncertain ; the object to attain 
 which an action is performed, and the subject 
 which the same action is likely to affect. The 
 relation of companionship, of instrumentality, 
 of similarity, and addition, will all, from the 
 same common principle, be expressed in it 
 likewise. 
 
 Lastly, the accusative will express every 
 object which is conceived in its property of 
 extension or duration, or which from the cir- 
 cumstances in which it occurs, necessarily 
 dwells on the mind, and repeats without inter- 
 ruption one simple and primary perception. 
 In this form, therefore, will occur the lines 
 which form the basis or parallel of a moving 
 body, the moving body itself, the point to 
 which it is moved, when considered as ex- 
 tended, as in the case of motion into ; and more 
 generally when it occurs to the eye again and 
 again as every object must do, towards which 
 we are progressively advancing ; so also the 
 substance in which any change is perceived to 
 take place, since change can never be perceived 
 without extension; so also the parallel line of 
 a substance conceived in extension either con- 
 tinuous or discrete ; so too any length of time 
 as well as of space. The subject of any afFec- 
 
30 GREEK PREPOSITIONS. 
 
 tion ; the point which in quitting we leave en- 
 tirely so as to vary that extension which is not 
 dwelt on while the object is relatively at rest; 
 and the occasional cause, or that, the presence 
 of which, without any implied communication, 
 produces a change in some other substance ; — 
 either the attention of the person acted on, 
 being supposed to be directed to the acting 
 cause, or a change in one substance necessarily 
 implying extension not only in itself, but also 
 in the parallel which affects it. 
 
 To adduce instances of all these uses would 
 require a voluminous work ; and they are ob- 
 vious and easy of access to every one in the 
 slightest degree acquainted with the Greek 
 language. But it may perhaps be worth re- 
 marking, that the laws of mind which sanction 
 this conjectural interpretation of the Greek 
 eases supply us also with three primary prin- 
 ciples under which all our pleasures and pains, 
 however complicated and abstract, may accu- 
 rately be classed. 
 
 Into the combinations, however, which these 
 cases form, the words denominated prepositions 
 perpetually enter. And, although our own lan- 
 guage by no means delights in elliptical ex- 
 pressions, which leave out the main word in 
 the sentence, that word which designates the 
 relation ; nor are English hearers so acute as 
 spontaneously to supply such deficiences with- 
 
GREEK PREPOSITIONS. 31 
 
 out guide or assistance ; the Greeks, if very 
 eminent scholars can be trusted, did possess 
 such a power, and consequently never hesi- 
 tated to employ words or combinations per- 
 fectly unmeaning, omitting the preposition 
 which gave light to the whole sentence, and 
 trusting for the interpretation to the ready sug- 
 gestion of their hearers. Whenever, in fact, an 
 oblique case occurs, which does not fall in with 
 the common grammatical syntax, a preposition 
 is said to be understood. And in this rational 
 creed we have all been brought up from our 
 infancies, although even the insertion of the 
 preposition could never account for the em- 
 ployment of the case ; and there was no more 
 reason why a particular case should occur in 
 those common and obvious forms which gram- 
 mar had already attempted to generalize, than 
 in others more rare and perplexing. What- 
 ever opinions are entertained respecting the 
 real significations of these inflexions, no philo- 
 sophical inquirer can doubt that some signifi- 
 cations do exist independently of any other 
 words ; and that prepositions are not requisite 
 to invest them with an adequate meaning. 
 
 Rejecting however this part of the office of 
 prepositions ; namely, the concealment of our 
 own ignorance by that most convenient of all 
 technical sophisms, the term subaudito ; it may 
 be worth while to frame some plausible hypo- 
 
32 GREEK PREPOSITIONS. 
 
 thesis respecting their real nature and use. 
 They constitute so very important an element 
 in the Greek language ; assume such a variety 
 of forms, and present such a labyrinth of signi- 
 fications to the student, that the absence of 
 any intelligible account of them, even in our 
 most popular grammars, is, perhaps, the most 
 singular defect in the whole history of philo- 
 logy. The conjecture vs^hich I have ventured 
 to offer on the nature of the cases, may perhaps 
 have appeared too abstract and refined to be 
 consistent with truth ; and it is probable that 
 the same objection may occur to the following 
 hypothesis. No one, however, can have stu- 
 died the Greek language, even superficially, 
 without observing a wonderful depth of meta- 
 physical knowledge exhibited in its structure ^. 
 In many parts it appears to have anticipated 
 some of our latest discoveries in the science of 
 mind. And whether we attribute it with a 
 Scotch philologist to an artificially constructed 
 system, or to gradual modifications, intro- 
 duced by very delicate and refined perceptions, 
 or even, as we may be tempted to think, when 
 the importance of this philosophical accuracy is 
 viewed in connection with the knowledge of 
 which it has been made the medium, to some 
 more providential circumstances ; there can be 
 
 ^ See Appendix (D). 
 
GREEK PREPOSITIONS. 33 
 
 very little doubt, if the a 'priori reasoning je- 
 commended by Locke is applicable to any lan- 
 guage, it is to the Greek. In the most remote 
 and depopulated countries at this very day, 
 relics of art and science are frequently dis- 
 covered, which irresistibly carry us back to 
 some anterior period of the world, as gigantic in 
 the growth of its intellectual powers, as in the 
 animal and vegetable creation. And a very 
 similar feeling is excited when we study the 
 language of the Greeks by the light of a 
 modern philosophy. What degree of know- 
 ledge was possessed by those who framed it, 
 we cannot tell. It might, indeed, have sprung 
 out of fortuitous contingencies. But it is very 
 much like finding a steam engine in the tumuli 
 of Siberia. 
 
 To return, however, to the Greek preposi- 
 tions. 
 
 If some former conjectures on the origin of 
 language are correct, it would seem that these 
 mysterious words must originally have stood 
 for some tangible, visible, and determinate 
 objects. However abstract they are at present, 
 they never could be abstract in their original 
 signification. Such a position is totally opposed 
 to every acknowledged principle of the human 
 mind. They must originally have been nouns 
 —and, without any very profound investigation, 
 it is very clear that they were nouns employed 
 
 D 
 
34 GREEK PREPOSITIONS. 
 
 to indicate place. Now, if we consider the 
 mode by which we designate locality, the ana- 
 logy will serve as a very easy clue to a probable 
 solution of the problem. To mark out the 
 place of a chair, or a table, or a book, or any 
 other object, what must be done ? Some other 
 external object must be taken with which it is 
 in juxtaposition ; as the chair near the win- 
 dow, the table by the bed, the book on the 
 shelf. And if this addition of a second object 
 would adequately point out the precise situa- 
 tion required, no prepositions would neces- 
 sarily exist in any language. But inasmuch 
 as A, though in juxtaposition with B, may still 
 admit of being placed in very many points and 
 positions, how shall the precise locality, the 
 one single point in which it exists, be accu- 
 rately pointed out? Surely by naming some 
 one part of B as an additional index, just in 
 the same manner as when we direct a letter to 
 a stranger, we annex the province to the 
 country, and the town to the province, and the 
 street to the town, and if it be necessary, the 
 house to the street; and it may be in some 
 cases even the part of the house to the number 
 which indicates it. The successive ideas gra- 
 dually limiting and narrowing the circle within 
 which the object is to be found, until no room 
 can exist for doubt or mistake. And as each 
 successive idea in this case is included in that 
 
GREEK PREPOSITIONS. 35 
 
 which preceded, and formed a part of it, so it 
 is probable that the same natural process was 
 observed by the original framers of the Greek 
 and other languages ; and that the second 
 index of locality was selected from a part of the 
 principal index — ^just in the same manner as we 
 now speak of horses abreast, or neck and neck, 
 of beating by a head, of being placed on the 
 back of a thing ; where the words breast, neck, 
 head, and back, are so many substantives in- 
 troduced to designate the exact position of the 
 first idea in its relation to the second, and are 
 themselves parts of the second. Now it is not 
 an improbable hypothesis, that the Greek pre- 
 positions were just as much nouns as these 
 words which we know to be nouns — that they 
 designated certain parts of things — that the 
 names were originally assigned to those parts 
 upon the same principles which have moulded 
 language in general, though, perhaps, not at 
 present to be traced in this particular instance. 
 Unless, indeed, in the oriental languages we 
 were to find that the roots of the Greek prepo- 
 sitions, were expressions for parts of the human 
 body, of agricultural instruments, or of any 
 other visible objects, which having assumed 
 the names first, would gradually, by the in- 
 stinctive process of abstraction, communicate 
 them to all analogous relations ; j ust as we call 
 the top of a mountain its head, and the front 
 
 D 2 
 
36 GREEK PREPOSITIONS. 
 
 of a house its face. And if we find that the 
 abstract notions thus attained, must, from the 
 very nature of things be limited, or at least ex- 
 tend to a certain number ; if we find that this 
 number exactly coincides with that of the 
 Greek prepositions; and if, on tracing out all 
 their various uses, we can, by the employment 
 of legitimate analogies, reduce them all under 
 these primary significations, then we may fairly 
 presume that the hypothesis, if not really cor- 
 rect, is at least a very singular coincidence 
 with a difficult and complicated cipher. 
 
 Now every visible object must present itself 
 to the eye in one of three forms — as a line, or a 
 superficies, or a solid — there is no other. And 
 we need not have recourse to geometry to prove 
 this. A line must evidently be composed of the 
 following parts, from its every essence as a suc- 
 cession of points. First, there must be a point 
 or part to commence from, and a point to end 
 with, and a space or line in the middle. With- 
 out divisibility into these portions it would be 
 no line. Here then are three primary sub- 
 stantive parts, to designate the locality of an 
 object. But a line may be either horizontal or 
 perpendicular. And those parts which are 
 merely considered as extremes in the prior 
 case, will naturally obtain diff'erent denomina- 
 tions in the second, because the physical per- 
 ceptions of ascending and descending, are 
 
GREEK PREPOSITIONS. 37 
 
 something very different from those produced 
 by traversing a plane. And lastly, the inter- 
 vening space between the extremities of the 
 line may assume three different appearances : 
 for it may either form a portion of the same 
 substance w^ith them, or it may be a different 
 substance connecting them together, as a chain 
 attached to two men; or it may be a chasm 
 unfilled up ; the line being formed by the 
 passage of the eye from point to point, and not 
 created by any positive continuity of body. 
 Besides these cases there seem to be no other 
 existing or conceivable. Let us suppose then, 
 merely for the sake of argument, that the word 
 irpo indicates that point or part of a line which 
 is first encountered ; that the other extremity 
 has no appropriate name, unless ott/ctw be at- 
 tached to it, but is expressed either by i-n^h or 
 vTto, according as the perspective presents it, as 
 it must do, in one of these lights. That ava 
 serves for the top portion of the perpendicular 
 line, Karcc for the lower, /xera for the intervening 
 part, when consubstantial with the extremes; 
 o-tv for the same when the medium is an ex- 
 traneous body ; and utcq for the same when it is 
 nothing but a space or interval. It may be 
 advisable to lay no stress on any conjectural 
 etymologies, though such as present themselves 
 evidently favour these definitions. 
 
 To proceed however to a superficies. This 
 
38 GREEK PREPOSITIONS. 
 
 also, from its very geometrical essence, coupled 
 with the necessary operations of our senses, 
 must give us the following parts. A plain 
 surface, a boundary line or circumference, a 
 space inclosed within that line, and another 
 space external to it. And it will also admit 
 of being divided by a line drawn through it. 
 Let us suppose that eVi stands for the plane, 
 -jript for the circumference, ev for the interior, 
 and €K for the exterior space, and ha for the 
 line which cuts the plane, dividing it into two 
 parts. Here also etymology is favourable. 
 
 Lastly, let us take a solid, and we shall ex- 
 tract from it the following portions. An upper 
 surface, an under surface, a side, and, where 
 the solid is hollowed out, two sides, a super- 
 ficies fronting us, and one in the rear. Besides 
 these a solid involves no primary essential 
 parts. 'T^rep then, by the very sound, expresses 
 the upper surface ; v^o, the under ; Trapa, the 
 side ; Scf^^), the two sides ; avr), the front ; and 
 what is singular, the rear is left apparently 
 without any representative, since it never could 
 be visible ; unless again we have recourse to 
 ott/o-o), of which the etymology is obvious. All 
 these respective words, if we judge by the ana- 
 logy of language, must originally have ex- 
 pressed corresponding portions of certain com- 
 mon and familiar objects, named not from any 
 consideration to their abstract relations, but 
 
GREEK PREPOSITIONS. 39 
 
 from some prior accidental circumstances ; and 
 their names must subsequently have been 
 transferred to the analogous parts of other 
 objects, until nothing was left to them but the 
 one common abstract relation, which to us has 
 been totally lost. 
 
 In this catalogue two prepositions have been 
 omitted, tt/o^ and hq. They appear to be com- 
 pounded of Tcpo and ev severally blended with the 
 particle <t€, which we find enter frequently into 
 the designation of place. And if we consider 
 that the local relation of one object to another 
 must be a relation either of motion or rest; and 
 that from these two cases we can only gain 
 three ideas, — those of a fixed, an mcreasing, 
 and a decreasing interval, we shall find reason 
 to believe that these were the original significa- 
 tions of 0t or Bex,, 6ev, and <re. The first being em- 
 ployed when the two objects were at rest ; the 
 second in the case of amotion; the third in 
 the case of approximation; or, to translate 
 them plainly, when A was either at B, or going 
 from it, or moving to it. However abstract 
 these little words, at, to, and from, with their 
 corresponding particles in Greek, may seem to 
 be, still it is evident, that the only ideas which 
 can legitimately be attached to them are those 
 which have here been mentioned, because no 
 others are or can be involved in any case, to 
 which they are applied, nfihq therefore will 
 
40 GREEK PREPOSITIONS. 
 
 mean to-before, or towards ; ci?, to-in, or in 
 English, into. 
 
 From these primary significations of the 
 Greek prepositions many others are deducible, 
 partly analogous, and partly comprehended 
 under the same term, from their uniform or 
 general coincidence in fact. To have traced 
 them regularly out into all their deflections 
 would have been a task far exceeding the 
 limits of the present observations. But after 
 rigidly applying the preceding hypothesis to 
 an infinite variety of cases, with a resolution to 
 admit none but natural and easy analogies, it 
 has been in no slight degree a confirmation of 
 the theory, a confirmation which the want of 
 time only has induced me to defer, that all the 
 prepositions without exception, and all their 
 innunerable meanings, have fallen naturally 
 into their places, and that even the stubborn 
 i^era, the most indocile and untractable of all, 
 which starts up at one time as withy at another 
 as afte7\ at another as change of place, at an- 
 other as participation, — even /^era has been re- 
 duced into order : 
 
 Succubuitque jugo, et lentas admisit habenas. 
 
 There are but two other questions respect- 
 ing these prepositions which require attention. 
 First, their adverbial uses, and secondly, their 
 indeclinable form. With regard to the former 
 
GREEK PREPOSITIONS. 41 
 
 it appears that they never deviate from their 
 strictly local meaning, even in the most abstract 
 combinations. And that the only difference 
 between their signification v\^hen joined with 
 cases, and their use when entering into com- 
 pound words, is derived from the expression or 
 implication of the main object, the relation of 
 which to some other object they serve to de- 
 note. Hapa with a noun, B, implies the position 
 of A at the side of B. n^pa, in composition, 
 when it signifies amiss, signifies the same A at 
 the side of some other point or standard, which 
 the sense of the context must supply. In one 
 case the second object is specified, in the 
 other it is merely suggested ; and so with all 
 the other adverbial uses of the several prepo- 
 sitions. 
 
 The question why these words, if nouns, are 
 not declinable, may be answered by referring 
 to a preceding hypothesis respecting the mean- 
 ing of those inflections of the noun which con- 
 stitute its declension. Both from the nature of 
 the thing, and the position of the words, it is 
 evident, that those parts of the second object 
 which denote the more immediate relation, the 
 more precise point of locality in which the first 
 stands to it, never dwell upon the mind as 
 separate ideas, susceptible of different phases, 
 but are introduced merely as adjuncts, and 
 attached to the main noun, much in the same 
 
42 GREEK PREPOSITIONS. 
 
 manner as any other initial elements of com- 
 pound words, which, though formed from sub- 
 stantives, never assume any variety of in- 
 flection. 
 
 The conjectures of etymologists are in gene- 
 ral so unsatisfactory, and in this case of so 
 little importance, if the hypothesis itself accord 
 with the facts, that I have cautiously abstained 
 from indulging them. But it may be worth 
 while to point out the frequent repetition of 
 the element ito, ita, in, and wep, in the composition 
 of these prefixes ; the recurrence of the same 
 element in the particle itep, and its acknow- 
 ledged significations of stability, firmness, cer- 
 tainty ; notions which would naturally be ex- 
 pressed by the analogy of a solid plane, and, 
 perhaps, of the surface of the earth. And if a 
 similar element with this latter meaning was 
 found in any cognate or original language con- 
 nected with the Greek, it would materially 
 support the conjecture. 'A[A,(pl evidently exists 
 in the pronoun aV^w, and ha in h<;, or two parts ; 
 jt/tera aud /^eVo^ arc also cognate terms ; a-lv, with- 
 out any great demand on the fancy, might be 
 connected with i^vylv; ^p^ we possess ourselves in 
 the shape of for, or, more obviously, as the 
 prow of a ship ; oitia-cc, out of sight, is also 
 clearly to be traced; and probably the few 
 which remain might find their original uses in 
 some t)riental tongue. 
 
INFLECTION OF NUMBER. 43 
 
 Upon these observations on the elements, 
 which constitute the substantive groups of 
 perceptions, v^hich in logic are termed the 
 objects of simple complex apprehension, the 
 consideration will naturally follow of the other 
 grammatical inflections in the noun, which de- 
 signate number and gender, and in adjectives 
 the notion of comparison. 
 
 I cannot, however, proceed in observations 
 so very abstract, without again excusing its ne- 
 cessity to those who are contented with what is 
 generally termed practical common sense ; and 
 conceive that the rules of syntax are a sufficient 
 explication of grammatical constructions ; just 
 as a common man perceives no mystery in the 
 descent of a heavy body, because the law of 
 gravitation accounts for it ; totally forgetting, 
 that the principles and the facts are in such 
 cases perfectly identical ; and that what we 
 term laws of nature, as well as laws of gram- 
 mar, are merely generalized expressions for the 
 phenomena themselves. In tracing, indeed, 
 the causes of things in the physical world, we 
 very soon reach a point beyond which we can- 
 not proceed. When we have resolved the 
 chain of consequences into their immediately 
 proximate links, the utmost end of human in- 
 quiry has been attained. But in the mysteries 
 of language we are quite sure that we are very 
 far removed from any such limits. They are 
 
44 INFLECTION OF NUMBER. 
 
 effects of which the causes must lie in the ope- 
 rations of the human mind ; and there we must 
 go for their explication. Unless, indeed, there 
 are any persons content to shelter their own 
 indolence under the plea of difficulty and inu- 
 tility; and anxious to attain a conviction that 
 the science of the human mind is unnecessary, 
 because, like every other branch of knowledge, 
 it demands patient and laborious investigation. 
 And if a different class of inquirers are willing 
 to acquiesce in a rough general statement of 
 facts, instead of a minute scrutiny of ultimate 
 principles, they should remember that such 
 coarse and inaccurate views of things, instead 
 of facilitating the acquisition of knowledge, in- 
 volve it in tenfold perplexities. The more 
 universal the principle, the more easy it is to 
 imbibe. When we are placed on the pinnacle 
 of the hill, we can easily find our way down ; 
 but are very likely to lose our course if we start 
 from any point short of the summit. And in 
 the science of mind, principles, when attained, 
 are, beyond all other rudiments of knowledge, 
 invaluable, from their unlimited application to 
 all the diverging operations into which they 
 run, whether moral, intellectual, or physical. 
 
 Starting, then, on this voyage of discovery, 
 without any presumption of final success, and 
 only resolute to persevere till we have ap- 
 proached as near as possible to the pole, we 
 
INFLECTION OF NUMBER. 45 
 
 may venture to propose the following conjec- 
 tures on the mode by which we acquire our 
 notion of number, principally with a view to 
 explain some apparent anomalies connected 
 with it in the Greek language. 
 
 When we use the expression, very common 
 in metaphysical writers,*that we possess in our 
 minds an idea of number, such an assertion is 
 very likely to mislead. Of the word number, 
 that is, of the sound, we have, indeed, a very 
 distinct perception : but of number abstract- 
 edly we can have no idea ; any more than we 
 can have an abstract idea of motion without a 
 moving body, or extension without some ex- 
 tended surface, or of time without a succession 
 of individual perceptions. 
 
 Let us, then, suppose a blind man suddenly 
 called into existence, and a sensation of warmth 
 excited in his mind : if, the moment this sensa- 
 tion ceased, his consciousness was extinguished, 
 it is evident that he would not have acquired 
 any idea of number. Again, suppose that the 
 same individual could exist in any number of 
 perceptions, however multitudinous and va- 
 rious, and yet those perceptions, or states of 
 mind, were perfectly insulated and detached 
 from each other by intervals of suspended con- 
 sciousness, it is equally evident in this case as 
 in the former, that no idea could be obtained 
 of plurality. Every perception would be to 
 
46 INFLECTION OF NUMBER. 
 
 such a man a point in existence without any 
 series connected with it either by memory or 
 anticipation. 
 
 Let us take a third case. Suppose by any 
 means the human eye, without any interrup- 
 tion from muscular sensation, could be made to 
 convey to the mind the idea of an unbroken 
 plane of colour, in which various shades and 
 gradations melted secretly and imperceptibly 
 into each other, as in the expanse of the hea- 
 vens, in which the deep blue of the centre 
 slides quietly into the light of the horizon ; 
 here again it seems certain, that although the 
 state of mind produced by such a transition 
 would in reality be very different, of that differ- 
 ence there would be no consciousness, and 
 consequently no notion of number. 
 
 Now, seeing in this last case the ease with 
 which the mind falls imperceptibly through 
 a series of states, but very slightly changed or 
 inflected, let us suppose a fourth case, in 
 which, being differently constituted, it would 
 pass with the same facility into states the most 
 opposite and remote from each other. And 
 here also it would seem, that although in each 
 sequence the feeling would be different, the 
 difference itself would be imperceptible. We 
 should no more perceive a distinction between 
 the colour green and the colour white, than be- 
 tween two imperceptible gradations of an aerial 
 
INFLECTION OF NUMBER. 47 
 
 tint. And that such would be the case is evi- 
 dent from experience. For we never think of 
 comparing, that is of observing distinctions 
 between ideas which fall naturally in with our 
 existing train of association. It requires effort, 
 and attention, and care, to discriminate at all ; 
 and if our minds were originally organized as 
 we can now mould them by habit, to slide 
 easily into any series of states however diffe- 
 rent, we should never have attained any notion 
 of distinct perceptions, and consequently none 
 of number. 
 
 The confirmation of this might be drawn out 
 to a great length, but it may be easily traced 
 by each individual for himself; I merely wish 
 to infer the conclusion that our whole per- 
 ception of distinction arises from that law of 
 mind which renders its operations the very 
 reverse of that just supposed ; which gives it a 
 tendency to pass on into one sequence in pre- 
 ference to another ; and which makes us sensi- 
 ble of a difference solely when this anticipated 
 sequence is checked or broken. 
 
 And this law may be reduced under two 
 heads. 
 
 First, a tendency of the mind, when uninter- 
 rupted by other external impressions, to repeat 
 and vibrate upon the same idea. 
 
 And secondly, a tendency to run into any 
 series of states, however dissimilar, which have 
 
48 INFLECTION OF NUMBER. 
 
 once been dovetailed into each other by re- 
 peated association. 
 
 So that we may arrive at a notion of dis- 
 tinction, and consequently of plurality, by two 
 methods. 
 
 Let a plane, of which half is painted black, 
 and half white, be placed before the eye of an 
 infant. If its mind can be fixed steadily on 
 the contemplation of one colour (and be it re- 
 membered, the idea of a plane is nothing but a 
 repetition of similar points of colour running 
 into each other, and generating lines by the 
 duration of the vibration or impression, after 
 the removal of the primary causes, just as a 
 stick on fire, when whirled round in the air, 
 describes a continuous circle) ; if, I say, the 
 child's mind, or perhaps its eye, can be made 
 to repeat consecutively, for a certain number of 
 times, the perception of this one colour, it will 
 acquire a tendency to continue the idea, and 
 when that tendency is checked by the occur- 
 rence of another colour and perception, it will 
 immediately become conscious of a difference ; 
 it will attain the notion of two colours, or two 
 states of mind. If the two colours were pre- 
 sented to the eye with such an interval be- 
 tween them as to allow the entire cessation 
 and dying away of one perception before the 
 other occurred, no distinction would be per- 
 ceived between them. And if any one doubt 
 
INFLECTION OF NUMBER. 49 
 
 this, let him endeavour to perceive the dis- 
 tinction between any two notes of music, or 
 any two shades of colour, or any two sensations 
 either of the touch, or the smell, or the palate, 
 without bringing them closely into juxtaposi- 
 tion, and he will ascertain its impossibility. 
 Such an attempt involves in fact a self-con- 
 tradiction, for it presupposes a perception of 
 relation, without the existence of relation ; the 
 discovery of a difference in a single term ; and 
 a comparison between two states of mind, when 
 we are conscious of only one. 
 
 Let us suppose then that a man born blind, 
 and brought into the world for the purpose of 
 the experiment, was made to taste an acid, 
 and immediately afterwards a sweet, or to have 
 his hand transposed, without interval, from a 
 fire to an ice pail, he would even by these two 
 perceptions, and the intervening sensation, 
 although instantly annihilated upon them, have 
 obtained a notion of difference, of number, of 
 succession, and, consequently, of time. 
 
 If his eyes were opened and permitted to 
 traverse a plane divided between two different 
 colours, excluding all intervention of muscular 
 sensation, he would in this case also attain to 
 the notion of two units, and two only, whatever 
 might be the extent of the coloured surfaces. 
 But the greater the extension of the first, the 
 stronger would be his perception of difference 
 
 E 
 
50 INFLECTION OF NUMBER. 
 
 in passing into the second ; since by a pri- 
 mary law of the mind our tendency to repeat a 
 perception increases with the repetition itself; 
 and the sensation of interruption is exactly 
 proportioned to the strength of the tendency. 
 If, therefore, a plane was divided between 
 three colours, each of equal extent, the per- 
 ception of difference at passing out of each, 
 would be equal likewise. And as the extent 
 was diminished, this perception would propor- 
 tionately fade away. Until, if reduced to a 
 single line, the mind would dwell on it so little 
 as scarcely to be sensible of any distinction be- 
 tween the two surfaces divided by it, and would 
 run them both into one. Thus to a savage a 
 man on horseback would present but a single 
 unit, since the eye would easily pass over the 
 faint shadows which marked the separation. 
 Let those shadows, however, be expanded into 
 any considerable space by the removal of 
 the man, and the substances would imme- 
 diately become two. The mind would be 
 checked and thrown back when it came to the 
 terminating lines, first of the man and then of 
 the space intervening. And as it is the sensa- 
 tion of this interruption which conveys to us 
 the notion of number, we should more naturally 
 say that the man and the horse were two, than 
 the man and the space between them. It need 
 not be observed that whether we discriminate 
 
INFLECTION OF NUMBER. 51 
 
 between two simple sensations, or between 
 two compound units, the several parts of which 
 have been blended together by frequent asso- 
 ciation, the process is precisely the same. The 
 perception of distinction, and consequently of 
 number, is in this latter case, as in the former, 
 produced by some check given to the mind 
 when running through its chain of ideas; either 
 by the absence of some expected link, or the 
 presence of one unexpected ; though, perhaps, 
 this second mode may be resolved into the 
 former. If we saw a buoy floating in the 
 water out of sight of land, and having accu- 
 rately collected together all the ideas which it 
 conveyed, were then to be shown another pre- 
 cisely the same, that is, producing no alteration 
 or break in the series into which we should 
 naturally fall, we should not conceive the exist- 
 ence of more than one : we should pronounce 
 them identical. But if the first buoy had pro- 
 duced five ideas, and the second suggested 
 only four, the check thus given to the mind in 
 its progress into the fifth, would make us 
 sensible of a difference, we should say there 
 were two buoys. And whether the link defi- 
 cient was part of the buoy, or a mark external 
 to it, but locally inseparable, as the shore near 
 it ; whether it occurred in the second, or was 
 perceived in the first, when recurred to from 
 the second, the process would be precisely the 
 
 E 2 
 
52 INFLECTION OF NUMBER. 
 
 same ; so that our notion of number may uni- 
 versally be resolved into that sensation pro- 
 duced by a check in our anticipations. It 
 cannot exist without the perception of simi- 
 larity, and of difference, nor without the notion 
 of time. And thus it is that we more readily 
 class together numerically groups of objects 
 which affect us in a similar manner, than those 
 which are very distinct ; that in counting, an 
 army for instance, we form our units from the 
 men rather than from the men and the horses. 
 
 Without remembering that there is a wide 
 distinction between existing differences and 
 existing numbers, and the mode by which we 
 become conscious of them, these observations 
 will appear frivolously and vexatiously minute, 
 and perhaps self evident.. But the vagueness 
 with which these operations of the mind are 
 usually described is productive of much mis- 
 chief. And when any phenomena are to be 
 explained, it is always desirable to do it by 
 ultimate principles, however simple and ab- 
 stract. The subject might be illustrated very 
 diffusely, and not without utility in connection 
 with the fine arts, very many of which involve 
 rules founded entirely on this process of the 
 mind. But at present we have only to apply 
 it to certain apparent anomalies in the nume- 
 rical inflections of the Greek language. 
 
 And first it may serve to explain why the 
 
INFLECTION OP NUMBER. 53 
 
 dual number is natural and philosophically 
 correct, since no comparison can take place 
 except between pairs. And the moment we 
 get beyond this point, if we wish to form any 
 positive idea of number, we are soon lost and 
 confused, and must be content with a vague 
 sense of unchecked progression, instead of any 
 clear and definite notion. And this is evident 
 to every one ; for we are perfectly unable to 
 count, except by creating artificial difierences, 
 and such differences as will singly and severally 
 imply all the units which have gone before, 
 without the necessity of actually retracing 
 them. Thus in counting the strokes of a clock, 
 either with our fingers, or by employing the 
 sounds which represent numbers : if we used 
 but one finger to beat with at each stroke, we 
 should immediately become perplexed ; but 
 by taking them in succession, each denotes the 
 exact place of the stroke which it represents; 
 and the words three, four, five, six, and the 
 rest, act in the same manner. We have our- 
 selves at each change a perception of two units 
 only ; and never could retain those which had 
 yjreceded, unless, with the mark of each as it 
 was added, we were able to combine a mark of 
 their place in the numerical order. 
 
 Secondly, the phenomenon of a neuter plural 
 nominative, and sometimes even a feminine or 
 masculine, coupled with a singular verb, may 
 
54 INFLECTION OF NUMBER. 
 
 be thus explained. For fixed inactive substan- 
 tives (and this is evidently the signification of 
 the neuter gender) naturally permit the mind 
 to form them, though distinguished by lines, 
 into units. We are checked much less in pass- 
 ing the eye from one house to another adjoining- 
 it, than in running through a number of sub- 
 stances, v^hich either are or have been seen in 
 motion. And hence, v^henever the neuter noun 
 in Greek expresses a living being, the verb 
 reappears in the plural. So also probably 
 v^hen the substances are not continuous, but 
 discrete, as in Homer : Ka.1 ^ lov-pa, o-eo-TjTre veSv, kcu 
 
 a-TVccprcc XeXvvroci. B. 135. 
 
 Thirdly, whenever a number of individuals 
 are massed together, as an army, or a crowd, 
 or where they are spoken of collectively in any 
 relation which does not necessarily imply an 
 interval between them, the whole compound 
 group thus formed by the same kind of easy 
 transition as before described, becomes a single 
 unit ; it will be expressed in the singular 
 number ; and whenever the predicate is the 
 result of their combined forces only, the verb 
 will be in the singular likewise. Whereas, if 
 the predicate be an attribute which is discern- 
 ible in each individual, the verb will be in the 
 plural. When an army repulses an enemy, the 
 verb is singular : when the same army are said 
 to be brave, the verb is plural. 
 
INFLECTION OF NUMBER. ^5 
 
 Fourthly, as dual nouns are sometimes 
 coupled with plitral verbs, on the principle that 
 the genus is included in the species ; so in 
 some peculiar instances the case is reversed, 
 and plural nouns, or more than two individuals, 
 as subjects, are connected with a dual verb. 
 And it need scarcely be observed, that in this 
 case the subjects, however numerous, are con- 
 sidered under two heads, and reducible into 
 two divisions. Thus Homer, addressing four 
 horses, speaks of them in the dual, because 
 they were coupled in pairs ; and so also of the 
 sailors who are ordered to furl the sails and 
 take down the mast, because in all probability 
 this operation required their separation into 
 two bodies. 
 
 To these principles several apparent excep- 
 tions may possibly occur : but by the admission 
 of exceptions generally so much mischief is 
 produced, in accustoming the mind to careless 
 views and inaccurate modes of thinking; and 
 the precision of the Greek language is so mi- 
 nutely and invariably philosophical, that every 
 effort should be made to reduce such anoma- 
 lies ; and even abandon, if unsuccessful, the 
 theories which refuse to include them. No 
 notion is more common among younger stu- 
 dents, than that words may be used indiscrimi- 
 nately ; that a possible sense is the real sense ; 
 and that the meaning is first to be imagined, 
 
56 INFLECTION OP GENDER. 
 
 and then the language distorted to suit it : a 
 principle which is obviously destructive of all 
 rational canons of interpretation, and all the 
 interest of philological studies ; but v^hich must 
 continue to prevail till v^^e remove as far as 
 possible the chapter of exceptions from our 
 grammars. 
 
 The nature of those inflections, v^hich in the 
 Greek and other languages distinguish the 
 genders of nouns, has been accurately ex- 
 plained by others ; and is itself so obvious as 
 to require but little illustration. 
 
 No one can have examined the structure of 
 the Greek noun, both substantive and adjective 
 (for in fact they are essentially the same, and, 
 as before observed, differ only from their acci- 
 dental position, as separate or connected v^ith 
 others), w^ithout perceiving that they are evi- 
 dently constructed from uninflected roots of 
 various terminations, united to the words which 
 have become known to us in a separate form, 
 as the article and pronoun. To trace this prin- 
 ciple through all the apparent irregularities of 
 the declensions, would require an elaborate 
 discussion : but to assert it in the most quali- 
 fied manner, it is at least highly probable ; and 
 it explains the cause of an essential distinction 
 on this head between modern and classical lan- 
 guages. The gender of a word is essentially 
 marked out by its composition in Latin and 
 
INFLECTION OF GENDER. 57 
 
 Greek. In the English, and many other 
 tongues, it is still left to the caprice or associa- 
 tion of each individual. That the freedom of 
 our own language in this respect gives great 
 scope for the fancy is very obvious. Poetry is 
 never so delighted as when embodying ab- 
 stractedness and inaction in living and breath- 
 ing motion ; and the whole energy of personifi- 
 cation frequently depends on the selection of 
 the gender. 
 
 But where the inflections of case were at all 
 regulated by the generic character, perplexity 
 would frequently be occasioned from the want 
 of an obvious classification under which to 
 refer nouns. And Grecian poetry sufl'ered less 
 than many others from the restrictions imposed 
 on its powers of animation and creation : en- 
 riched as it was by mythology with a multi- 
 tude of invisible agents already invested with 
 reality. 
 
 The principle, however, which guided the 
 Greeks in their generical distinction of nouns 
 was precisely the same as that which influences 
 ourselves and all men in their daily personifi- 
 cations of inanimate objects. Whether fixed 
 in the language, or left open to variation, the 
 rule was still laid down in the laws of our 
 natural associations. We discriminate every 
 moment between dead, motionless, and passive 
 substances, and such as are instinct with action 
 
58 INFLECTION OF GExVDER. 
 
 and animation ; and the peculiar ideas and 
 associations which arise from the relation of 
 the sexes immediately suggests a subordinate 
 division of active beings. The existence of 
 any quality or analogy primarily perceived in a 
 man, or a v^oman, or an inanimate substance, 
 would classify the object which contained it as 
 masculine, feminine, or neuter. As we apply 
 the word horse to distinguish other substances, 
 in which the notion of strength, magnitude, 
 and consequent coarseness is involved, as a 
 horse chesnut, a horse radish, and even a horse 
 laugh ; so the word which primarily denoted, 
 perhaps, an individual man, was applied to 
 everything endued with the characteristics of a 
 man. And thus too in the other genders, 
 strength, activity, preeminence, duration, and 
 other derivative attributes, were designated by 
 the masculine noun, or what is become to us 
 the pronoun, oq. Weakness, fragility, the object 
 of affection, the principle of production and 
 nutrition, dependence, subjection, in short, 
 every feeling and idea which flow from the 
 relation in which woman stands to man, would 
 be in like manner indicated by the name of 
 woman. And the analogy of all languages is 
 curiously consistent in applying these two 
 sounds respectively, as they exist in Greek, to 
 the several qualities thus naturally coexistent. 
 Lastly, under the head of neuter nouns would 
 
INFLECTION OF GENDER. 59 
 
 be placed everything considered as merely 
 passive, as useless, sometimes as contemptible, 
 but in a greater degree than when the feminine 
 attribute is annexed ; and in general every ob- 
 ject to which no definite notion was attached, 
 to reduce it under either of the previous heads. 
 Three usages of the neuter gender may be 
 mentioned, which are curious not from their 
 peculiar employment in Greek, but from their 
 frequent appearance in other languages. First, 
 its signification of a diminutive, even when the 
 primary notion is a living being ; and this 
 ^sometimes in contempt, sometimes affection- 
 ately. A nurse talking to her baby, and De- 
 mosthenes thundering against ^schines, would 
 equally divest their subject of all pretensions 
 to vitality and action. The phenomenon, per- 
 haps, may be thus accounted for. The notion 
 of power, which we derive entirely from the 
 perception of principles of motion, naturally 
 involves that of dignity and value ; and the 
 opposite characters will be easily designated 
 by the opposite analogies. And if we look into 
 the constitution of our nature, we may find an 
 explanation of our tendency to neutralize the 
 objects of affection; either in the frequent con- 
 nection remarked by Burke between minute- 
 ness, delicacy, beauty, and consequently love, 
 and the natural association of diminutiveness 
 with inactivity and helplessness; or perhaps 
 
m INFLECTION OF GENDER. 
 
 still more deeply within us, in that mysterious 
 attraction which the littleness and dependence 
 of an object exercises on our sympathy ; a 
 principle which, if fully developed, might illus- 
 trate the most singular prodigy in the whole 
 history of man ; his proneness to create objects 
 for worship out of brutes and inanimate sub- 
 stances. 
 
 A second usage of the neuter gender, found 
 principally in the tragedians, and designating 
 the first person, is curious, but obviously natu- 
 ral; since the parts of the body are, separately, 
 inactive, and would be considered in that light 
 when referred to as signs of the existence of 
 the man himself, or the essential part of his 
 nature. 
 
 A third case in which it is employed, is to 
 express animated beings, particularly in the 
 plural number : ra tcAtj tSv AaKe^aiiAonav, instead of 
 the magistrates, as we speak of the court of St. 
 James, or the Ottoman Porte. And generally 
 in such instances something is predicated of an 
 office, or relation, rather than of the individuals 
 who fill it ; and the relation itself is naturally 
 expressed by some one of its circumstances, 
 without necessarily implying action. And this 
 particularly holds good where the office is per- 
 manent, and its occupiers shifting. The cabi- 
 net instead of ministers, and the pulpit instead 
 of the preacher, are metaphors common on the 
 
INFLECTION OF GENDER. 61 
 
 same principle. Other cases might be ad- 
 duced of a similar kind : but they easily occur 
 to the student, and may probably be all ac- 
 counted for and reduced to some general prin- 
 ciples. The anomaly of nouns masculine in 
 one number, and neuter in the other, might 
 probably be brought under the observations 
 before made, respecting the inflections of num- 
 ber. Thus in Latin to those who believed in 
 the fixedness and stability of the firmament, 
 ccelum v^ould be the natural word to express it; 
 while those who conceived the heavenly bodies 
 to be set and whirled round in a number of re- 
 volving spheres, ccelos would be the appropriate 
 term. And the same conjecture might, per- 
 haps, be applied to many instances in Greek of 
 a similar nature. For it never can be repeated 
 too often, that grammatical rules are not 
 founded on caprice or whim, but, like all other 
 effects of human agency, spring from some 
 cause in the human mind. And to be satisfied 
 with a technical word as an explanation of a 
 phenomenon, without tracing the fact back to 
 its primary source, is a much more mischievous 
 proceeding, than to tax the imagination for a 
 clue which, if it does not satisfy the reason, 
 may at least exercise the understanding. Every 
 thing must have a cause, and to ascertain these 
 causes is the business of philosophy, whether 
 working on the physical operations of nature, or 
 
62 FORMATION OF SUBSTANTIVES. 
 
 the inflections of a dead language. The process 
 in both is precisely the same ; and it is to pre- 
 pare the mind for the one pursuit, that we ex- 
 ercise it in the other. 
 
 The inflections which have hitherto been con- 
 sidered of gender, number, and case, are com- 
 mon to both adjectives and substantives; and 
 in grammatical language they agree in them ; 
 or, as it is expressed with greater impropriety, 
 the substantive governs the adjective. The 
 real cause of this apparent harmony will be 
 obvious, when we observe the mode in which 
 the same word performs the transition from 
 one class of terms to the other. In English, 
 having no inflections of the kind, no such con- 
 cordance takes place; and whether we gain by 
 the variety of terminations, which this want 
 permits to us, or lose by an occasional obscu- 
 rity, may be a difficult question to solve. 
 
 If a previous observation on the distinction 
 between substantives and adjectives be correct, 
 if they are respectively the same word repre- 
 senting the same quality, or group of qualities, 
 and assume an independent or connective cha- 
 racter solely by the relative position which 
 they occupy in the mind ; a single horse in 
 shafts, to use a common illustration, indicating 
 the substantive, and the same horse in a team 
 acting as an adjective, it is a natural question, 
 by what means they have become so widely 
 
FORMATION OF SUBSTANTIVES. 63 
 
 separated, both in the Greek and in other lan- 
 guages, as to form two distinct classes of 
 words. And the process is the following. It 
 has before been remarked, and it is too well 
 known to require remark, that from the want 
 of a separate sign for each single object, we 
 are frequently obliged to denote an individual 
 by the word which designates a class ; that is, 
 by some quality which is in no way peculiar to 
 it, but is found in many others. We wish to 
 describe the thing which contains our pen and 
 ink, and we call it a stand. To reduce the 
 number of objects, within which it may be 
 found, still farther, we add to stand, a word ex- 
 pressive of certain other qualities, ink ; and 
 thus the doubt is limited to those things only 
 in which these two circumstances are found 
 united. Precisely in the same manner, when 
 we wish to direct persons in discovering the 
 author of an action who is known to ourselves, 
 but not to them, we describe him first perhaps 
 as an acquaintance ; then, if this is not suf- 
 ficient, as an acquaintance living in such a 
 country ; and, if this is not enough, as an ac- 
 quaintance living in such a country, and of 
 such and such character. And thus we con- 
 tinue adding circumstance to circumstance, 
 until a certain number are accumulated, which 
 collectively are applicable to but one indi- 
 vidual. It may also be observed, that in form- 
 
64 FORMATION OF SUBSTANTIVES. 
 
 ing these primary distinctive combinations, 
 which, if we choose to employ logical terms, 
 we may call species composed of genera and 
 differentiae, we seldom employ more than two 
 members, as ink-stand, lamp-oil, writing-book, 
 tea-chest, and the like ; and that the rest of 
 the designation is performed by the article, 
 with some other adjuncts, not immediately 
 blended with the principal or generic index. 
 As this process is perfectly natural and in- 
 stinctive, it is not surprising to find it employed 
 by the Greeks : and the first difference which 
 they selected in order to limit the number of 
 individuals, within which the object to be spe- 
 cified was to be sought, was the word which to 
 us has become the pronoun and the article, but 
 which assuredly in its primary use signified a 
 man, a woman, or some passive inactive object. 
 The union of the radical quality with the quali- 
 ties designated by the pronoun substantive 
 (which at that time it must be remembered 
 was no pronoun at all) served to indicate the 
 individual in which alone these two series of 
 elements were found. And when a farther 
 specification was required, the same process 
 was repeated ; and the same pronoun was not 
 again subjoined, but prefixed, in order to mark 
 out the exact individual. This prefix to us is 
 the article, and its peculiar uses will occur 
 hereafter. Thus the word kux would express 
 
FORMATION OF SUBSTANTIVES. 65 
 
 the first object, A, which . affected the mind 
 with a feehng of beauty. And if no second 
 object occurred which produced the same re- 
 sult, this A would not require the annexation 
 of any other word to the former to specify it 
 accurately; kocx would be a substantive. But 
 when many objects of the same kind occurred, 
 as B, C, and T>, A would require to be pointed 
 out by some feature of difference. This differ- 
 ence the Greeks took from the real or imaginary 
 gender. If the qualities denoted by the mas- 
 culine, for instance, were found united with 
 the radical quality in one individual only, no 
 farther difference would be necessary ; and Kaxl^, 
 a beautiful man, would be as much a substan- 
 tive as /cax was before. But when the number of 
 things in which this union occurred increased, 
 then some new attribute, or even many, were 
 selected to single out the unit intended. And 
 without such an attribute /caXo^ became an ad- 
 jective. The faet is still more obvious in the 
 proper names of the Greek language, which 
 still retain their adjective form, and substan- 
 tive use, even in some cases without the article. 
 The terminations tas, tudo, and others in the 
 Latm language, and ity, 7iess, hood, and the like 
 in English, are employed in the same manner, 
 in the same relative place, and for the same 
 purpose, as the generic noun in Greek. As Ka,\, 
 a quality found in many things, comes to de- 
 
66 FORMATION OF SUBSTANTIVES. 
 
 note an individual by the annexation of o?, a 
 man, to it ; so quantus becomes quantitas ; simi- 
 lis, similitudo ; chaste, chastity ; good, goodness ; 
 man, manhood. The nature of these latter ter- 
 minations, without a knowledge of the Saxon 
 and Celtic languages, it is impossible to trace. 
 There is no reason to suppose that they neces- 
 sarily classified universal terms under the 
 heads of gender ; and perhaps they served to 
 form some more subordinate division, as the 
 feminine termination in Greek is used with the 
 article to denote a particular science, or virtue, 
 as species of feminine things, generally recog- 
 nised and familiar. 
 
 But the principle of formation is evidently 
 the same ; and no difficulty can arise from the 
 variety of terminations in the Greek noun, 
 whether oq, tj, ov, uq, ev;, tq, 1, 73?, '/jp, or a, since all these 
 are reducible to primitive forms of the same 
 generic or pronominal substantive. These con- 
 jectures, if correct, will assign the philosophical 
 cause of the general agreement in combination 
 between the substantive and adjective, in gen- 
 der and number, and their universal agreement 
 in case. Since the sign of the main object 
 described enters necessarily, from the structure 
 of the language, into every member of the de- 
 scription ; and as all these members are simul- 
 taneously coexistent in it, it will appear in 
 each under the same form and position, and be 
 
FORMATION OF SUBSTANTIVES. 67 
 
 placed in the same case. But as the associa- 
 tions which influence the designation of gender 
 and number may vary in a moment, in these 
 the agreement is sometimes interrupted ; mas- 
 culine adjectives being connected with neuter 
 nouns, and singular nouns with plural adjec- 
 tives. The same principle will also remove 
 the necessity of considering those numerous 
 expressions as elliptical in which the adjective 
 occurs alone, or merely coupled with the ar- 
 ticle. The termination which denotes the gen- 
 der being in fact if not a substantive, at least 
 the sign of that abstract substantive for which, 
 in English, we employ the word thing, and 
 being already included in the same term with 
 the adjective quality. It is quite as dangerous 
 to trust a grammarian with an ellipsis, as a 
 stranger with the power of filling up blank 
 drafts, or a rhetorician with rhetorical figures 
 and poetical licenses. And it would confer a 
 great benefit on all young students if a short 
 piece of composition were shown them in 
 which advantage was taken of all these ima- 
 ginary forms, and their consequent absurdity 
 demonstrated. In the infancy then of the 
 Greek language, and probably of every lan- 
 guage, the same primitive sound performed the 
 reciprocal duties of adjective and substantive. 
 In the English extensive traces of this structure 
 still remain ; but in Greek they are nearly obli- 
 
 F 2 
 
68 FORMATION OF SUBSTANTIVES. 
 
 terated : and a line of demarcation has been 
 drawn between the two, which in many cases 
 it is impossible to pass. A tendency to a 
 similar discrimination is visible in our own 
 tongue : and the mode in which it is effected 
 is well deserving consideration. By simply 
 casting the eye over the pages of a Lexicon, 
 even the most moderate scholar will intuitively 
 class the nouns which occur as substantives or 
 adjectives. Some he will be assured are never 
 employed as adjectives, others never as sub- 
 stantives ; and the terminations will serve as 
 his guides. The inflections, therefore, of the 
 primary idea, are the instruments by which 
 this distinction has been eff'ected. And they 
 have eff'ected it by means of the additional 
 ideas denoted by them; and which, by their 
 very essence, erect the elements to which they 
 are annexed into independent nouns, or reduce 
 them into epithets. 
 
 The elements which inflect the radical so as 
 to form a necessary substantive are /^, o-, and t. 
 They constitute very prominent and important 
 features in the Greek vocabulary, and their 
 general significations are familiar to every 
 student. The words in which they occur are 
 usually supposed to be derived from the pre- 
 terperfect passive of the verb; but it would 
 perhaps be more correct to consider the two as 
 collateral stalks from the same root. Each 
 
FORMATION OF SUBSTANTIVES. 69 
 
 involves certain ideas which may be termed 
 verbal, but the latter contain some v^hich are 
 not included in the former, and the addition of 
 which in the participle of the preterperfect 
 enables the noun to reappear as an adjective. 
 noiVa and iremiriiA.evov arc two very distinct words. 
 The former must always be a substantive from 
 the notion implied in its penultimate /* : the 
 latter contains the same ideas precisely, but 
 something else besides, indicated by the redu- 
 plication and the termination €voq ; and this ter- 
 mination enables the idea conveyed in ^o/ry/Aa to 
 become attributive. 
 
 This will become very obvious if we consider 
 that a substantive word expresses any idea or 
 number of ideas, which, either by perception 
 or abstraction, we insulate and detach ; that it 
 does this by the sign of one single quality con- 
 tained in that number (for in compound sub- 
 stantives the initial elements, though united 
 in the same word, are evidently as much 
 adjectives as if separately marked by the 
 generic inflection), and that this quality or 
 attribute, whether generic or any other, natu- 
 rally infers the existence of some substratum 
 which supports it; since, from our incapacity 
 of seeing into solids, the frequent change of 
 external shape and colour in objects whose 
 materials are yet unaltered, and, in fact, from 
 the very relation which exists between the 
 
70 FORMATION OF SUBSTANTIVES. 
 
 sentient substance of our mind and its states 
 and objects of perception, qualities are to us as 
 so many coats or coverings spread over some 
 supporting material, w^hich v^e can only sup- 
 pose to exist, but can never perceive ; and 
 which stands in the same relation to the quali- 
 ties which invest it, as canvass to paint, or a 
 table to its covering 
 
 To trace minutely the origin of this uni- 
 versally received notion is not necessary. But 
 it is evidently an inference drawn from analo- 
 gous perceptions, and not a matter of fact pro- 
 position. It is obvious, however, that a word to 
 indicate this substance, or matter, or vXv}, divested 
 as it is in our minds of all qualities, must be 
 divested of all qualities itself. And the diffi- 
 culty, if not impossibility of framing such a 
 word, induced the Greeks and others to indicate 
 its existence only by attributes as abstract as 
 possible, though by no means so abstract as 
 was required — those namely of gender. Still 
 the union of the particular quality with the ge- 
 neral quality of gender, served but as a sign for 
 the substance in which they were supposed to 
 exist. The word which contained them re- 
 mained essentially an adjective, and became a 
 substantive solely by its accidental appropria- 
 tion. How then was the essential character of 
 a substantive given to an attributive word ? 
 
 It is quite obvious that this can be done 
 
FORMATION OF SUBSTANTIVES. 71 
 
 solely by abstracting the quality; that is, by 
 placing it in any light in which it must at least 
 seize the attention apart from the notion of any 
 other substance. And in such a distinct and 
 separate form a quality does appear when we 
 see any change taking place in an object, by 
 the operation of some efficient cause. When 
 we minutely examine the ideas conveyed to 
 the mind by any individual case of action and 
 passion, we shall find that they are all re- 
 solvable into the perception of some new quality 
 appearing in a substance; gradually and con- 
 comitantly with the presence of some other 
 substance. This quality, by its novelty, irre- 
 sistibly seizes the attention, detaches itself 
 from the others to which it has just been an- 
 nexed, assumes a substantive form, and, from 
 the necessary analogy of motion to a definite 
 point, is designated in Greek by the letter i^, 
 which almost invariably possesses this precise 
 signification. And no equally correct and phi- 
 losophical mode could be devised by which 
 to express an indefinite substance as recipient 
 of some new quality, than by that new quality 
 itself with the additional notion of its recent 
 annexation. In the same manner if we ob- 
 serve the state of our mind when watching a 
 process by which some change is to be eff'ected, 
 that is, some new attribute to be attached to a 
 substance, we may observe, that the attribute 
 
72 FORMATION OF SUBSTANTIVES. 
 
 itself exclusively occupies our thoughts, and 
 that there is a perpetual tendency to anticipate 
 it. Expectation precedes reality : and an ex- 
 perimental c J mist is always ready to declare 
 that the changes in his liquids have taken place 
 long before they actually occur. And hovs^ is 
 this process denoted in Greek ? by the sign of 
 the same quality in a substantive form, and 
 inflected by the insertion of the letter o-, w^hich 
 in the verb, and in many other combinations, is 
 demonstrably the index of inceptiveness, of 
 tendency to, of something future and expected. 
 And for the <t in Greek the r is substituted in 
 Latin, when the process of production is de- 
 scribed, just as in the inceptive verbs ; what is 
 o-eiw in one language, becomes rio in the other. 
 Natiira, factiira, qucestiira, are the analogous 
 forms for viwria-K;, itorrjo-i^, ^-/^ttjctk ; as factuyno cor- 
 responds to 'KOifia-ela, aud faCturUS to 'KOi-qaccv. 
 
 Lastly, a similar method was observed in in- 
 dicating the cause of a change ; and the agent 
 was denoted by the sign of the quality pro- 
 duced, inflected by the letter t, which occurs in 
 the particle ^ev, and many other words in the 
 signification of motion from a place. 
 
 It will readily be allowed, that these conjec- 
 tures on the precise and original meaning of 
 those nouns which indicate the agent, the action, 
 and the thing done, are very abstract and re- 
 mote fr..m their present obvious employment. 
 
FORMATION OF SUBSTANTIVES. 73 
 
 Their exact accordance, however, with the 
 natural process of the mind, their uniformly 
 substantive character, their subsequent inflec- 
 tion by genders w^ithout altering this character, 
 and the abstract character of the thing signified, 
 demonstrated by the universal applicability of 
 the sign, are all strongly conducive to the same 
 result. That the first inflection by /^ should 
 be susceptible of three genders, that by t of 
 two, and that by o- of the feminine only, is per- 
 fectly consistent with what might be expected. 
 The result of an action is of course open to any 
 farther notion of activity, passiveness, or neu- 
 trality. The agent can never be considered in 
 the contradictory light of a non-agent. And 
 the action in all probability is marked uni- 
 versally by the feminine from the metaphorical 
 notion of production. 
 
 These principles might, if pursued farther, 
 supply rules for tracing the gradual ramifica- 
 tions of words in Greek through their several 
 parent stocks. But at present it is sufficient 
 to observe, that the three forms above men- 
 tioned are the only essentially necessary sub- 
 stantives in the language ; that all others 
 become so accidentally, by the combination of 
 qualities which infer the material on which 
 they are engrafted. And that these combina- 
 tions are sometimes expressed at full length, as 
 avBfccTtcx;, a mail ; sometimes formed by an adjec- 
 
74 INFLECTIONS OF ADJECTIVES. 
 
 tive and the article, as rl Ka\Lv, sometimes ellip- 
 tically implied, when the gender has been by use 
 appropriated to particular substantives, as -jroXt- 
 tiich, aper-ri, OY avl§la, — but iu all of thcsc, whcthcr a 
 noun is understood or not, there is no ex- 
 pression corresponding to the English word 
 thing, which constitutes our universal substan- 
 tive. The termination of the gender and the 
 article, indicate in themselves nothing but 
 qualities, and the substratum is only implied. — 
 In the cases of verbal nouns above mentioned, 
 the genders also represent qualities, but the 
 attribute on which they are grafted is essen- 
 tially and necessarily a substantive. 
 
 Without a minute knowledge of the lan- 
 guages from which the Greek shot out, it is 
 impossible to ascertain with precision the sig- 
 nification of those inflections which distinguish 
 the varieties of adjectives. Whether the cases 
 are formed by the addition of any extraneous 
 element to the radical term, or merely by a 
 change of pronunciation, may perhaps be 
 doubted. But with respect to the penulti- 
 mate inflections of adjectives the question is 
 hardly admissible, and we may feel assured, 
 that if the same process was pursued in classi- 
 cal philology which has been applied by 
 Home Tooke to English etymology, the result 
 would be in the same manner a still more 
 minute analysis of forms apparently most 
 
INFLECTIONS OF ADJECTIVES. 75 
 
 simple. A priori reasoning is in this case very 
 fallacious, for the modifications of which an 
 attribute is susceptible, seem to be limited by 
 no natural principles. And the frequent sub- 
 stitution of one mode for another, whether 
 adopted for the preservation of euphony, or 
 suggested by existing affinities, leaves us with- 
 out any fixed standard to apply as a test to 
 conjectures. An hypothesis therefore can do 
 little more than attempt a rough classification 
 of adjectives under a few general heads, with- 
 out presuming an exact conformity in their 
 various uses. 
 
 First then we may describe objects either 
 by their positive or negative qualities — and 
 the latter class of terms are used in two senses, 
 to express sometimes the absence of a quality, 
 sometimes the presence of an opposite. In 
 the former sense it is evident, that they must 
 always maintain an attributive form, and be 
 combined with some substantive either ex- 
 pressed or implied, since negation abstractedly 
 can never act as real existence. 
 
 Positive qualities may be subdivided into 
 those in which existence only, and those in 
 which motion is implied. The latter class 
 comprehending all verbal adjectives. — These 
 again are either active or passive, and the ac- 
 tive either transitive or intransitive. And as 
 action which passes on to some subject termi- 
 
76 INFLECTIONS OF ADJECTIVES. 
 
 nates either in a mere affection of it, or in pro- 
 duction, the Greeks appear to designate the 
 former case by the termination po<;, the latter by 
 f^av' — ^"Kd^epoq, a man acting mischievously ; (po^e^lq, a 
 man causing flight ; Xiy.fipoi;, causing hunger ; hxeplq, 
 working treacherously : but jvuiaccv, producing; yvS- 
 (A.a,i — ^pd^i^uv, producing reasons; or (ppcc—ixevjixuv, pro- 
 ducing pity. The termination iko<;, appears also 
 to be employed occasionally in an active but 
 intransitive sense, to signify a tendency to an 
 action or state ; just as we employ the word 
 like in English. — But its verbal nature seems 
 to depend not upon itself but on the word to 
 which it is subjoined. 
 
 Of passive verbal adjectives there are three 
 obvious terminations to?, reo?, and o-;^o^ — The first 
 applied to the subject of an action finished, the 
 last to a subject capable of being acted on, 
 and the intermediate reo; conveying the notion 
 of duty — corresponding respectively to the tus, 
 bilis, and ndus, of the Latin language. But the 
 active and passive senses are so frequently ex- 
 changed, that the line of distinction cannot be 
 drawn with any accuracy. 
 
 Another class of words, essentially adjectives, 
 are those which are employed as substitutes 
 for second appended nouns in the genitive or 
 dative case. And they seem divisible into two 
 heads, according as they are employed for the 
 one or the other. The termination tvo? appa- 
 
INFLECTIONS OF ADJECTIVES. 77 
 
 rently corresponding with the former, and that 
 of io^ with the latter — pihvov eXcclov, naturally im- 
 plies oil made of, or from roses ; pohela raivla,, a fillet 
 made in roses. — And thus perhaps ,o? comes to 
 denote the attribute of place and also of time ; 
 xapa5f*o?, opviq a bird in, or inhabiting a torrent ; 
 r^iraio<; Terapraiot;, in OY on thc third or fourth day. 
 
 The last class which may be mentioned, are 
 those which annex to the quality some index 
 of its relative quantity. 
 
 Where the existence of the quality merely 
 in a slight degree is expressed, in English by 
 the termination 3/, or in vulgar language ish, as 
 fiery, handsomish, reddish, watery, the Greeks 
 seem to employ </<f? and eo^ — where its exist- 
 ence in a great and generally offensive degree 
 is to be marked, as the Latins used osus the 
 Greeks have recourse to ^Stj? — where the stand- 
 ard of comparison is a number of individuals, 
 the penultimate is distinguished either by a re- 
 duplication, or a combination of consonants 
 which arrest the pronunciation. — And that this 
 is a perfectly natural mode of expressing great- 
 ness of degree, is evident from its perpetual 
 employment by ignorant people and children ; 
 who either repeat the word, or lay an emphasis 
 on its penultimate, whenever such a notion is 
 to be conveyed. 
 
 And lastly, if the standard is a single unit, 
 the particle rep is introduced, or the word is in- 
 
78 INFLECTIONS OF ADJECTIVES. 
 
 fleeted in ^wj/.— -This latter termination is probably 
 Oriental, and the former Celtic. Of rep, which 
 occurs in erepo<;, and in many words of northern 
 origin, the signification of difference seems ob- 
 vious — and probably it might be traced into 
 the numerals with the same sense to direct its 
 application. It is evident that our perception 
 of equality, like that of similarity, is obtained 
 by a very complex mental process : that is, by 
 finding no link deficient in a chain of antici- 
 pated ideas, which we have previously ac- 
 quired by experience. In the case of simi- 
 larity the links appear to be severally different; 
 in that of equality they are repetitions of the 
 same primary perception. — One house is like 
 another when the majority of ideas called up 
 by each are respectively the same — a line is 
 equal to a line, when the number of sensations 
 produced by the muscular action of the eye in 
 traversing them is checked in each at the same 
 point ; and as these sensations are very slight, 
 and cannot accurately be marked and retained 
 in the memory, hence it is scarcely possible to 
 attain a notion of equality without bringing the 
 two lines into a close parallel. — In the case 
 also of simple qualities, as tastes, smells, co- 
 lours, and the like, the perception of equality 
 in degree is only possible when either to the 
 senses or the memory the two sensations are 
 brought close together; and the transition 
 
INFLECTIONS OF ADJECTIVES. 79 
 
 from one to the other is marked only by a dif- 
 ference in place or time. If no such difference 
 exists, the several perceptions melt into one. 
 An increase, however, or diminution of degree, 
 instantly constitutes a difference, in the same 
 manner as any two perceptions, however dissi- 
 milar, are distinguished in the mind ; that is, 
 by breaking and checking the associations. — 
 And hence the propriety of marking such a 
 change of degree by the sign of difference 
 attached to the index of the quality which ex- 
 ists positively in both units. This also may 
 explain the use of the comparative degree with 
 the genitive case in Greek, and the adjective in 
 Latin — since the transition from the first per- 
 ception to the second would be marked ex- 
 actly in the same manner as motion from any 
 one place to another. And the distinction is 
 still farther marked by the particle n, which 
 answers the same purpose as the English or, 
 and seems to draw a line of demarcation be- 
 tween the two objects divided by it. A vul- 
 garism in the English language is very analo- 
 gous to this use of 'ri,—this man is better nor he, 
 is not an uncommon provincial expression. — 
 The phrases la-oq kou, ceque, ac, and the like, are 
 to be explained in the same manner; by the 
 tendency which the mind has in the case 
 of equality to run the two perceptions into 
 one. 
 
80 INFLECTIONS OF ADJECTIVES. 
 
 Two phenomena seem worthy of attention 
 in the use of the comparative degree. 
 
 First, the absence in very many languages, 
 perhaps in all, of any peculiar inflection to de- 
 note diminution in degree, as well as increase; 
 and secondly, the essential character of attri- 
 butiveness which the comparative bears. 
 
 The first may perhaps be accounted for by 
 the natural tendency of the mind to fix and 
 dwell upon the object which produces the 
 strongest impression ; and consequently to 
 make it the first term in the enunciation of the 
 comparison. 
 
 The second evidently arises from the fact, 
 that although a combination of qualities, or 
 even a single quality, as in the designation of 
 the Deity by the term Good, may be sufficient 
 to mark out a substantive to which they are 
 exclusively appropriate, no such accurate de- 
 signation can be performed by a word which 
 essentially presupposes the existence of the 
 same qualities in other objects — unless those 
 objects were confined to two, and the term of 
 comparison marked the increase of degree in- 
 stead of merely distinction. Neither of which 
 suppositions exists in reality. 
 
 Under the head of substantives and adjec- 
 tives are evidently to be ranged two classes of 
 terms usually described as adverbs and pro- 
 nouns. 
 
THE ADVERB. 81 
 
 Of the adverb it is obvious, that it is the ad- 
 jective of the adjective ; and is employed to 
 express any modification in the quality of the 
 substantive. Hence it occurs principally in 
 words descriptive of magnitude or degree. — 
 And in English it is possessed of a peculiar 
 termination, ly, which is usually supposed to 
 be a corruption from like. In Greek it appears 
 generally to be the case of an adjective — the 
 dative when the first quality is supposed to be 
 contained in the second ; just as substances in 
 ordinary languages are said to be contained in 
 figure, and any object to be comprised and 
 surrounded as it were by the quality which ap- 
 pears on its superficies — the accusative, proba- 
 bly in its old termination w?, and sometimes 
 also in the singular, when the model or fashion 
 of the action or affection is considered — after 
 such a pattern, according to such a method, a 
 signification which is inherent in the accusative, 
 and requires no ellipsis of a preposition to be 
 supplied. The neuter plural of the adjective, 
 which is sometimes termed an adverb, is per- 
 haps universally to be reduced under the ordi- 
 nary principles of grammatical arrangement, as 
 a noun with the substantive suppressed. And 
 the same observations may be applied to many 
 of the adverbial inflections in the Latin lan- 
 guage. That the adverb, properly so called, as 
 marking an attribute of an attribute, must be 
 
82 THE PRONOUN. 
 
 originally an adjective, is obvious from its very 
 nature. And if this proper and peculiar sense 
 of the word be adhered to, we must at once 
 exclude from the class to which it is applied, 
 every term which is not reducible to such an 
 origin ; and either leave the infinite number of 
 signs, whose etymology is unknown, in a heap 
 by themselves, with an acknowledgement of 
 our ignorance, or bring them back by careful 
 inquiry into their proper places as substan- 
 tives, verbs, participles, pronouns, or which is 
 more common, combinations of several of these 
 terms. 
 
 The next element of language to be ex- 
 amined, is the pronoun ; and there is no one 
 more interesting and curious, both from its own 
 metaphysical character, and the accurate ana- 
 logy which on this point runs through every 
 variety of dialect. 
 
 It is perfectly evident, that all our percep- 
 tions are conversant with individuals; and it 
 may also be assumed, that when we reason with 
 abstract qualities, though the signs of them are 
 applicable to classes of things, in reality our 
 notions are still confined to the individual qua- 
 lity. As these individuals resemble one ano- 
 ther in many points, it is necessary to have the 
 means of distinguishing them ; that is, of fixing 
 on the unit to which attention is to be directed, 
 and calling the mind of the observer from all 
 
THE PRONOUN. 83 
 
 others of the same class. Either one circum- 
 stance, or a combination of circumstances, is 
 therefore to be selected, which is appropriate 
 and peculiar to it, and constitutes its difference. 
 The selection of this difference, and the extrac- 
 tion of what is logically termed the essence of a 
 compound substance, is the result of a very 
 long, complicated, and ingenious process of 
 reasoning, in nine hundred cases out of a 
 thousand. In many it is utterly impossible. 
 And if the constitution of things did not supply 
 us with the means of obtaining external differ- 
 ences by which to mark out individuals, instead 
 of analyzing their intrinsic elements, we should 
 soon be reduced to extreme perplexity and 
 ignorance. Place, therefore, is the natural, 
 obvious, and general mode of distinguishing 
 individuals ; and accordingly we find it em- 
 ployed every day, where no single and distinct 
 sign has been previously appropriated : and no- 
 where is its usage more conspicuous than in 
 the original names of persons and families in 
 our own language. It is also obvious, that we 
 require these differences of place for the pur- 
 pose of communication only. They are the 
 external aids by which we draw the attention 
 of others to a point on which our own attention 
 is already fixed. And lastly, it is clear that 
 the place or second sign which we select, must 
 be known and permanent ; and that it may be 
 
 G 2 
 
84 THE PRONOUN. 
 
 assumed either within ourselves, or external to 
 ourselves. The tower of London, Demosthenes 
 the Athenian, the tree on the lake, are in- 
 stances of the second kind. And the former 
 kind is exhibited in the pronouns. 
 
 Now, as in every communication there must 
 be at least two individuals, and individuals of 
 the same species, it will be necessary to dis- 
 tinguish them ; that is, as before observed, the 
 speaker must have some means of fixing the 
 attention of the hearer on the one which he 
 wishes to designate. Any description of its 
 qualities is perfectly useless, since the object 
 itself is placed before the senses. A single 
 gesture, or motion of the hand towards the 
 point to be specified, would be the instinctive 
 and easy method. That such gestures are per- 
 fectly natural, is obvious from our frequent 
 employment of them. We have an instinctive 
 tendency to touch every thing which occupies 
 our senses. A child grasps at every object of 
 sight. A clown begins to handle every thing 
 within his reach. And if such familiarity was 
 not checked by the decorum of society, we 
 should seize by the buttonhole, or clap upon 
 the back, every person to whom we addressed 
 ourselves, just as we put our hands to our 
 breast when speaking of ourselves, or point to 
 our own bodies when prohibited from speaking. 
 No illustration is necessary to show that a 
 
THE PRONOUN. 85 
 
 motion of the hand to our own body is instinc- 
 tive and natural when speaking of ourselves; 
 and a motion of the same hand from our body 
 when our attention is directed to a person ad- 
 dressed. But motion, and muscular motion 
 especially, affects the voice. And certain mo- 
 difications of sound respectively accompany 
 gestures, as necessarily and universally as they 
 are attached to the sensations of pleasure and 
 pain. The pantomime of the deaf and dumb 
 will show at once that the organs of speech are 
 moulded differently when the arm is drawn to, 
 and thrown from the body. And we may ob- 
 serve the same fact in ourselves, when we de- 
 scribe any action which admits of these gestures 
 being employed with sufficient violence to mark 
 the distinction. That the letter i^, which enters 
 largely into the first person, is descriptive of 
 a point to which motion is directed, and the 
 letter o- of motion in progress, is evident from 
 the whole structure of the Greek language. 
 The frequent confusion which occurs there be- 
 tween the elementary signs of the second and 
 third person, that is, between a-, and r, is also 
 perfectly natural on this hypothesis, since very 
 little distinction of gesture is perceptible when 
 pointing to an object external to ourselves, 
 whether it be an individual addressed, or a 
 third separate substance. That the notions of 
 agency and passiveness are not the primary 
 
86 THE PRONOUN. 
 
 significations of the pronouns, as Gebelin sup- 
 poses, is evident from their frequent intermix- 
 ture in the formation of the verb. That some- 
 thing very abstract, and of universal applicability 
 was intended, is clear from the absence even of 
 generic distinctions in the signs of the first and 
 second person. And it is also obvious in vs^hat 
 manner modifications of motion, ideas of action 
 and passion, and numerical symbols, would na- 
 turally be expressed by similar elements with 
 the pronouns of the first, second, and third 
 person. If these pronouns were originally 
 formed in the manner presupposed (and the 
 exact coincidence of these facts with the hypo- 
 thesis at least renders it probable), it might be 
 a subject of curiosity to inquire by what ana- 
 logy these pronouns in so many languages are 
 irregularly formed ; and by what laws they 
 became in Greek and Latin the basis of inflection 
 in the cases, and are still inflected in our own 
 and other tongues, when all other nouns remain 
 unchanged? But without a very extensive 
 acquaintance with other languages, the search 
 would be impracticable. 
 
 The pronouns, therefore, of the first and se- 
 cond person, were signs which severally marked 
 the individual speaker, and the one addressed ; 
 and were equivalent to, and coincident with 
 certain gestures which designated the place of 
 the person specified. But as in all cases of 
 
THE PRONOUN. 87 
 
 communication, two points at the least are fixed 
 and known, we may employ them as local 
 marks to distinguish any units which are ex- 
 ternal to them. And as all gestures which in- 
 dicate place are made by the arms, and their 
 number and position divide the whole field of 
 vision clearly into two portions, we find that 
 there are two pronouns in language, this and 
 that, ovroq and eVeTVo^, hie and ilk, to designate an 
 individual by its position relatively to ourselves. 
 To these, the Latins have added iste, to mark 
 locality in relation to the second person, and 
 %vro<; in Greek, as opposed to oSe, sometimes 
 seems to bear the same meaning. 
 
 It need not be observed that eV and o-i/ are 
 necessarily substantives, from their being seve- 
 rally applicable to one individual. Whereas 
 ovroq and iKehoq arc either adjectives or substan- 
 tives, according to circumstances — substantives, 
 when the union of the locality and the gender 
 are sufficient to mark out the individual — adjec- 
 tives, when some other attribute requires to be 
 annexed. And hence many differences be- 
 tween the Greek and English idioms ; in the 
 latter of which the pronoun this or that can 
 rarely be a substantive, from its abstract and 
 simply local character. 
 
 When an individual has once been fixed and 
 described, it is frequently requisite to recall 
 the attention of the hearer to it. And as the 
 
88 THE PRONOUN. 
 
 repetition of the description would be tedious 
 and unnecessary, it is superseded by the em- 
 ployment of a sign which suggests it again to 
 the mind. To do this, such a sign must neces- 
 sarily be very abstract, since it is to be appli- 
 cable to an infinite variety of subjects ; and yet 
 not so abstract as to embarrass the hearer in 
 throwing him back upon former associations. 
 And since the process is to be performed by 
 association, it is evident that the sign must 
 be something previously connected with the 
 individual alluded to. It must be part of 
 some former chain of ideas, or by the laws 
 of our mind it never could serve to recall 
 that chain. A sign thus constituted is found 
 in all languages, in the words which indicate 
 gender. He, she, and it, is, ea, id, tq, ri, %, are 
 signs perfectly indistinct in themselves, but 
 which nevertheless are known to indicate some 
 one individual in the speaker's mind, because 
 words are never used without some such proto- 
 type. The hearer is immediately thrown upon 
 his search to find out the object intended. That 
 it is something known to himself is certain, 
 since we never voluntarily employ words in- 
 capable of suggesting the ideas required. That 
 it has been lately subjected to his observation, 
 is natural ; since otherwise the speaker would 
 not anticipate its recurrence. And the index 
 of the gender, while it limits the field of search, 
 
THE PRONOUN. 89^ 
 
 serves also to recall the notions with which it 
 was previously connected. The same distinc- 
 tion which was made between the adjective 
 and substantive use of tvroq and eKem?, may be 
 applied to the pronoun 'k. In Homer, in the 
 dialogues of Plato, and in many other Greek- 
 works, it is a substantive. In its adjective form 
 it generally appears as the article. Sometimes 
 the individual has been already described ; and 
 its use is purely relative. Sometimes it has 
 been presented to the speaker only, and a fuller 
 definition is appended to it for the benefit of 
 the hearer. Sometimes no such definition is 
 added, and it remains as r)q, leaving on the 
 hearer an impression that some one individual 
 was intended, without any knowledge of the 
 one specified. Hence the use of ri^ as an in- 
 terrogative. A question, in fact, being merely 
 the expression of an opinion in a doubtful form. 
 The principal differences between the Greek 
 and English idioms, in the use of the article, 
 arise from the generic character of the Greek 
 pronoun, and the absence of such a character 
 in the English the. Hence the article in Greek 
 may be used with prepositions, with adverbs, 
 with participles, and with adjectives ; where, 
 in English, no such phrase is admissible, ex- 
 cept under peculiar circumstances, where the 
 nature of the adjunct acts as the generic in- 
 flections. Thus we speak of the goody the 
 
90 THE PRONOUN. 
 
 vicious, the indolent; but never of the wooden, 
 the watery, or the metallic. The nature of the 
 quality in the former case restricts the ap- 
 plication of the predicate, in the latter not. 
 
 To enter however into an analysis of all the 
 different usages of this important word would 
 be vain without adducing examples, and al- 
 most superfluous after the minute examina- 
 tion which it has received from others. Its 
 nature as connected with the structure of the 
 language is all that is here alluded to, and in 
 this it is curious to observe its analogous for- 
 mation from the pronoun in such a number of 
 lanofuas^es. 
 
 Of the relative 'k it is scarcely necessary to 
 observe, that it is the same pronoun. And as in 
 Latin the relative qui is perhaps formed from /<«*, 
 and is; so in Greek the connecting particle re is 
 found attached to k and other pronouns in the 
 Homeric dialect, though subsequently it was 
 dropped. And here also there is a close analogy 
 to be observed in the English language, in the 
 evident formation of who from he, — The pro- 
 nominal forms in fact, from their very nature, 
 are those which have run out into every variety 
 of dialect with less alteration in their structure 
 than any other words, and they are of themselves 
 a sufficient proof of the identity of the source 
 from which language originally sprung. 
 
 As the pronoun 'k is in itself a relative, that 
 
THE PRONOUN. 91 
 
 is, as it signifies something which has been 
 previously mentioned or suggested; so it enters 
 into many relative adjectives. Its dative case ro), 
 seems to be the origin of ro7o<;, in perfect consist- 
 ency v^ith the general use of that case to indi- 
 cate the standard of comparison. From the 
 same element, with a-, inserted as the sign of 
 number, roVo^ appears to have sprung. Though 
 perhaps both -roVo^and iVo? may be merely a redu- 
 plication of the same pronoun to express iden- 
 tity. And from the genitive of toVo?, too-oSto? prob- 
 ably arose, in which perhaps the genitive of an 
 adjective of number serves as the measure of 
 quantity, from the close analogy between the 
 two notions. The immense number of particles, 
 which in combination with prepositions and 
 conjunctions it has thrown off under various 
 forms, deserve well the attention of the philo- 
 logist. — They have created no little confusion 
 in the principles of Greek composition, by 
 giving shelter to technical explanations of diffi- 
 culties, which are only supposed to be ex- 
 plained because we are ignorant of the means 
 resorted to. It is considered a satisfactory so- 
 lution of a phrase in which the Latin subjunc- 
 tive occurs, to say that it is governed by ut. 
 Whereas, as Home Tooke has clearly shown, 
 ut is merely the pronoun on. "iva., %a, ioq, perhaps 
 the hypothetical particle e*, certainly eVe*, e^ce*, 
 and many others, are nothing but the pronoun 
 
92 THE VERB. 
 
 in particular cases and combinations : and can 
 no more govern a subjunctive mood, than the 
 original noun itself. Uoq appears to have been 
 the adjective form of r)q — and the numerous 
 compounds of o? and r)^, or v^hat are usually 
 called the definite and indefinite pronouns, 
 seem to be employed only as more precise 
 substitutes for the original rk, to mark out a 
 particular individual, but to indicate at the 
 same time that he had not been previously de- 
 scribed. From oq also is formed avrog, or in a 
 literal translation, he again; from the particle av, 
 — 'Avroq, consequently, is used sometimes as the 
 mere pronoun — but more frequently the repeti- 
 tion of the individual serves to throw the mind 
 back upon it v^hen passing on to others of the 
 same species, and thus distinguishes and sepa- 
 rates it from the rest. Hence a man is said to 
 act avrU, when he does it by himself, without 
 assistance, or spontaneously, without compul- 
 sion. So also when an action is unproductive 
 of any results, avruq is employed ; and the addi- 
 tion of the article serves to indicate identity, on 
 the same principle as A is said to be the image 
 of B, or, when the likeness is still more strong, 
 to be B himself. 
 
 The most important part, however, of the 
 Greek language is the verb, and its considera- 
 tion necessarily involves an inquiry into the 
 nature of those complex forms of speech which 
 
NATURE OF PROPOSITIONS. 93 
 
 are denominated propositions. The following 
 conjectures are hazarded with great diffidence, 
 but may perhaps throw light on the subject. 
 It does not appear then that a logical proposi- 
 tion has anything to do with the comparison 
 of ideas. — It asserts the connection of two 
 ideas either simple or complex, but not their 
 similarity or difference. — Either the term com- 
 parison does not mean here what it does else- 
 where, or it is most improperly applied. — To 
 suppose that the assertion Ccesar is dead, infers 
 any resemblance between Caesar and Death, is 
 a notion too absurd to be refuted. When we 
 pronounce upon the truth or falsehood of a 
 statement, then, indeed, there is a comparison 
 implied, between the ideas suggested by the 
 words and those excited by the external ob- 
 ject, — and a statement is true when the two 
 trains coincide, and false when they differ. 
 But in itself a logical proposition expresses 
 simply the connection of two ideas in the 
 mind, or two qualities in a subject. 
 
 When these propositions are expressed in 
 words, and not before, it is very important to 
 observe that they branch out into two kinds, 
 one in which the second term, or predicate, is 
 included in the sign which stands for the sub- 
 ject, the other where it is merely connected. 
 
 The first class are necessary, the second con- 
 tingent. — And as no one simple state of mind 
 
94 NATURE OF PROPOSITIONS. 
 
 can include another, and all our perceptions 
 when analyzed are simple, it is evident, that to 
 us there can be no necessary proposition respect- 
 ing matters of fact, no proposition that is, which 
 we may not conceive to be otherwise. — And it 
 is also obvious, that whenever the predicate is 
 included in the subject, this inclusion takes 
 place by virtue of the signs which we employ. 
 The word which stands as the subject indicating 
 a number of ideas, one of which separately, and 
 under a different symbol is placed in the pre- 
 dicate. Again, as demonstration is nothing 
 but the showing that the two terms of a propo- 
 sition are necessarily blended, that is, that the 
 second cannot be detached from the first ; and 
 as this takes place only where the hypothesis 
 of language has included one in the other, it is 
 most important to remember that we can have 
 no demonstration except by words, and of 
 words. — And it is equally important to see, 
 that the impossibility of doubt in necessary 
 matter, and its possibility in contingent matter, 
 arises from the physical formation of our 
 minds, which cannot be in two different states 
 at the same moment; though they have the 
 power of checking themselves in following up 
 a train of perceptions, and breaking off any one 
 link from another. — That very many necessary 
 or identical propositions do not appear so to 
 ourselves is very certain. And in these cases 
 
NATURE OF PROPOSITIONS. 95 
 
 the difference in the signs which we employ, 
 acts in the same manner as a difference in the 
 ideas themselves. — But it may safely be as- 
 serted, that wherever a proposition is presented 
 to us which we cannot conceive to be other- 
 wise, its predicate has already been included 
 in the subject. Its validity depends upon hy- 
 pothesis : and its inference is purely verbal. — 
 The very momentous and practical conse- 
 quences which flow from this principle, if cor- 
 rect, are well worthy of serious consideration. 
 At present it is merely stated in reference to 
 certain grammatical constructions. 
 
 The difference between the formation of a 
 complex substantive, consisting of many con- 
 nected perceptions, and the enunciation of a 
 proposition, is simply this ; that in the latter 
 case the mind breaks off one link in the analy- 
 sis, and fixes upon it in a separate form. — And 
 the principles which influence us in this selec- 
 tion, are the same which draw and attract our 
 attention to any quality or substance amidst a 
 group. 
 
 These principles are three, novelty, plea- 
 sure, and pain. — And hence we do not express 
 our connected perceptions in the shape of pro- 
 positions, except where the predicate or quality 
 fixed on is new, or agreeable, or painful. The 
 exclamations of a solitary individual talking to 
 himself, are confined to these three heads. — 
 
96 NATURE OF PROPOSITIONS. 
 
 Hence it is that we never take the trouble to 
 enunciate propositions which are self-evident, 
 or obviously identical ; or which are known to 
 the hearer. Hence it is that a traveller de- 
 scribes the objects which he sees, solely in re- 
 lation to his own feelings of surprise or satis- 
 faction. And the want of interest sometimes 
 perceptible in narratives, and of clearness in the 
 arguments of the most learned and gifted men, 
 arises from the suppression of many proposi- 
 tions which are too familiar to be noticed by 
 the speaker, though perhaps unknown to the 
 hearer. The illustration of the fact may be 
 left to the personal observation of every one. 
 
 As the predicate then, or quality which ar- 
 rests the attention, is thus detached from the 
 subject or group with which it is perceived to 
 be connected, the separation is marked by a 
 break in the sense and the language, that is, 
 by the apodosis. But as such a break might 
 seem to imply two distinct and unconnected 
 perceptions, it would be perfectly natural, when 
 the second term was enunciated, to subjoin 
 some element which would refer to the former, 
 — and would in fact repeat it. Such an element 
 is found in the pronoun.^ — A savage expressing 
 his perceptions does naturally talk in this man- 
 ner — '' sun, it, rising ;" '' this wine, it sweet;" 
 ** Sambo, me good;" or, *' man, he wicked." — 
 And this appears to be the explanation of that 
 
FORMATION OF THE VERB. 97 
 
 curious phenomenon in the Greek language, the 
 formation of the verb from the attribute and pro- 
 noun ; a fact which having been led to anticipate 
 a priori, I have since found fully developed in 
 the celebrated work of Valckenaer. In addition 
 to many other difficulties explained by this 
 analysis of the verb, it may be worth while to 
 notice the natural formation of the passive 
 voice from the accusative of the pronoun ; the 
 various significations of the middle voice, ac- 
 cording as the termination stands for the dative 
 or accusative ; the analogy between the impe- 
 rative of the Greek and those English vul- 
 garisms, in which the word it is attached even 
 to intransitive verbs ; the identity of the verbs 
 in /xi, and the Latin forms with those in w ; the 
 formation of inflections in the verb, to designate 
 number and person, from inflections previously 
 existing in the pronoun ; and the falsity of that 
 hypothesis which makes verbs impersonal, or 
 conceives that in any case an ellipsis of the 
 nominative or subject could exist, such an 
 ellipsis being totally opposed to all the princi- 
 ples of human reason. 
 
 From the explanation here attempted of the 
 formation of propositions, it is easy to see the 
 mode in which the verb became susceptible of 
 various accidental modifications. In its original 
 and essential form it is nothing but an attribute 
 coupled with the pronoun. When the person 
 
 H 
 
98 FORMATION OF THE VERB. 
 
 indicated by the pronoun has been previously 
 designated, it forms in itself a whole and per- 
 fect proposition. When this has not been 
 done, a fuller description is subjoined in the 
 subject or nominative case. And even the 
 position of the pronoun at the end of the attri- 
 bute, as if its annexation v^ere an afterthought, 
 seems illustrative of the fact. In all this there 
 is no necessary notion of time, of action, of 
 passion, or even of being; except inasmuch 
 as the existence both of the subject and predi- 
 cate, at least in the mind of the speaker, is 
 essentially comprised in the enunciation of the 
 terms. The verb of being is in no way neces- 
 sary to a proposition. It is perfectly super- 
 fluous, as expressive of existence ; and perfectly 
 accidental, as indicative of time : and however 
 abstract its meaning is at present, originally it 
 must have been as complex as any other verb at 
 present existing. And yet that the notions of 
 time, of agency and of passiveness, and many 
 other modes of being, are attached to the verb, 
 is indisputable. And the solution of this phe- 
 nomenon lies in the nature of the principles 
 which influence our selection of the predicate, 
 by fixing our attention on some one quality 
 rather than on others. A fact, that is, a con- 
 junction of two perceptions, must be, in some 
 degree, new in order to surprise or aff'ect us. 
 The second term must be one unexpected; con- 
 
FORMATION OF THE VERB. 99 
 
 sequently it must be capable of existing sepa- 
 rately from the subject. The subject must 
 have before been perceived without it ; and 
 thus the notion of time is immediately conveyed. 
 So also action and passion are very generally 
 indicated by the verb ; because motion of any 
 kind, change of any kind, naturally arrests the 
 attention. To remark that the fields were 
 green in the middle of summer would be ridi- 
 culous ; but every one makes the statement at 
 the beginning of spring. No one but an idiot 
 would think it necessary to inform another that 
 the Alps were in the same place as they were 
 a thousand years ago ; but of a body capable 
 of motion, this might naturally be thought wor- 
 thy of mention. The observations made by a 
 child, or indeed by any one, when surveying 
 a landscape or travelling, would form the best 
 illustration of these principles. And since every 
 fact or conjunction may ultimately be conceived 
 capable of separation (for even identical pro- 
 positions rest upon hypotheses and arbitrary as- 
 sumptions, which may be denied if we choose), 
 it is perfectly natural to mark the time at which 
 they appear, and to make a distinction between 
 the state of a thing at the present moment, and 
 the state which it either has been in before, or 
 may be in hereafter. The precise moment at 
 which a king dies, or an earthquake occurs, is 
 noted down upon the same principle. And 
 
 h2 
 
100 INFLECTIONS OF TIME. 
 
 thus even necessary propositions are expressed 
 with reference to the present time only, though 
 the connection appears indissoluble when the 
 hypothesis, which supports it, is allowed. 
 
 It is also perfectly natural that the notion of 
 time should be attached to the predicate or 
 verb, because it is this which is conceived to 
 be the variable term ; and which suggests the 
 notion itself. Hence in the Greek, the Latin, 
 and the modern languages of Europe (for it is 
 not safe to extend the observation beyond the 
 little field which is open to our own personal 
 knowledge), the verb is inflected to mark the 
 time. And what is very well worth notice, it 
 is generally inflected so as to make but two 
 divisions, the past and the present. And the 
 future is either formed by the insertion of some 
 other verb, or is evidently to be ranged under 
 the present ; both from its uses, the nature of 
 our perceptions, and the analogous character of 
 its inflections — / strike, I struck, I will strike, 
 rvitrco, ervTtroy, rv^pco. Thc susccptibility of a qua- 
 lity, or tendency to it, seems to be spoken of as 
 a quality existing at present ; and, / am going 
 to strike, is the literal translation of rvrpco ; the 
 letter o- answering the same purpose as the je 
 vais of the French, and our own verb of motion. 
 And if we consider what is meant by the time 
 of an action, it may perhaps lead to a conjec- 
 ture respecting the nature of the Greek aug- 
 
GREEK AUGMENT. 101 
 
 ment. As we mark the place of an object in 
 space by the object to which it adjoins, so we 
 mark the place of a fact in the order of time 
 by the fact or perceptions which are imme- 
 diately connected with it. We cannot indeed 
 date as present time a perception which is 
 actually present to the mind, because it is itself 
 the standard of all other time. But we do say 
 that such perceptions take place in present 
 time as closely follow on some one other per- 
 ception, which recurring again and again at 
 very little intervals, seems to be, in some de- 
 gree, a fixed and permanent mark. This mark, 
 however, is a floating point, and varies every 
 moment in its relative position. And as the 
 enunciation of our perceptions implies their 
 present existence in the mind, it is perfectly 
 unnecessary to subjoin the date, when such 
 present existence is intended. But a fact is 
 said to take place in past time, when it was 
 closely conjoined with some one other fact not 
 at present existing. In this case the second 
 fact is a fixed and immutable point, known and 
 determinate, but described or not, according to 
 circumstances. And such a point might very 
 naturally be marked by the old pronoun of the 
 third person, e, which indicates some object or 
 conjunction not now present, but known and 
 consequently past. Whether the hypothesis is 
 worth notice may be left for the consideration 
 
102 SCHEME OF TENSES. 
 
 of others. But it is curious that the Latins 
 who want the augment, want also the definite 
 article, which came from the pronoun ; and 
 employ the pronoun itself but rarely and re- 
 luctantly. 
 
 The primary division then of the inflections 
 of the Greek verb, would naturally be formed 
 from present and past time. But under each of 
 these heads occur several other inflections or 
 tenses, which are wanting in our own and in 
 some other languages ; and it is important to 
 settle their precise signification and number. 
 And this may perhaps be done. by referring to 
 the respective analogies of the Greek adjectives 
 and the Greek cases. 
 
 For if the verb is merely an adjective, or 
 attribute in a difi'erent combination, it will 
 naturally be capable of the same modification. 
 Thus both the verb and the adjective may be 
 classed under the three heads of active, passive, 
 and neuter. And the subordinative modes of 
 action or production; the relative connection 
 between the quality and the subject ; the ten- 
 dency to a state, and the repeated perception 
 of a quality, are marked in each by distinct and 
 frequently similar inflections — thus eco seems to 
 express merely the existence of the quality — 
 w an action — ^w, oa, and lua, production as ex- 
 hibited severally in the creation of a substance, 
 or of a quality; and of a quality accordingly as 
 
SCHEME OF TENSES. 103 
 
 it is intimately or superficially connected with 
 the subject — the termination aBu would seem 
 analogous to that of the superlative degree, and 
 might bear perhaps a similar sense — <r6<« indi- 
 cates the tendency, and ctkco the same tendency 
 in an incipient state — and very nearly the same 
 notions are to be found in the inflections of the 
 adjectives. 
 
 A different classification of adjectives may 
 therefore supply us an analogous distribution 
 of the tenses, and this has been attempted 
 already. For we may describe an object by its 
 negative or positive qualities, and its positive 
 qualities may be either such as really exist, or 
 such as are capable of existing — and those 
 which really exist, may be subdivided into ab- 
 solute and relative, accordingly as they are con- 
 sidered in themselves, or with reference to a 
 prior state of the object. We may describe the 
 night as starless, or starry, or clouded, or ter- 
 rible — a fruit as unripe, or ripe, or ripened, or 
 edible — and a tree as lifeless, or green, or 
 blighted, or shady. It need scarcely be ob- 
 served that the negation of a quality naturally 
 implies its supposed prior connection, either in 
 the same or a similar object ; and thus we may 
 describe an object, either by a quality which it 
 no longer possesses, which is past and gone, or 
 by a quality which it possesses absolutely, or 
 by one which has been superinduced on it, or 
 
*^ 104 SCHEME OF TENSES. 
 
 by one which it is capable of possessing or 
 producing. And when we turn to the Greek 
 tenses, we shall find these four divisions accu- 
 rately marked by distinct inflections; and where 
 the nature of things permits, a perfect set of in- 
 flections severally assigned both to past time 
 and to present. The analogy of the Greek cases 
 would lead to a similar conclusion. For the re- 
 currence of the mind to a qualit}?" no longer per- 
 ceived in an object, its perception of an existing 
 quality, and its anticipation of one to come, are 
 three operations of the mind precisely similar 
 to those by which an object is perceived in its 
 unity and totality, in its extension, and in its 
 prospective position. On a quality no longer 
 perceived, we cast as it were but one glance, 
 and no more — on one which is obvious to our 
 senses, we dwell with reiterated perceptions — 
 on one which is to come, we look forward as 
 through a vista, with a sort of suspense and 
 pause, anticipating its appearance. And these 
 are the notions which have been assigned to 
 the Greek cases, and which at least will serve 
 to explain all the curious phenomena con- 
 nected with them. The observation of the 
 student will be sufficient to show him that some 
 such analogy does exist between the tenses 
 and the cases from some of their respective 
 combinations in instances of dates. The im- 
 perfect tenses running with the accusative, and 
 
Active. 
 
 PRESE 
 
 r Absolutely. 
 
 flMPERFECT.< r ' a v 
 
 ^Relatively. { Ta!?",' -* 
 
 TVTrrio, verbero, 
 I am beating. 
 
 verberavi, 
 beaten. 
 
 <1 <l Inceptive. 
 
 5 
 
 ^Perfect. 
 
 Tv\l/(t), verberabo, 
 I am going to beat. 
 
 r Absolutely. 
 ^Imperfect. < 
 
 (_ Relatively. 
 
 > 
 
 ^ <{ Inceptive. 
 
 Perfect. 
 
 ^ rvTTTb), verberem, 
 ^ I may be beating. 
 5 TSTv^u), verberaverim, 
 ( I may be having beaten. 
 
 ^ Tv-ipeia, verberaturus sim, 
 X I may be going to beat. 
 
 ^ Tv\l/it), verberaverim, 
 i^ I may finish beating. 
 
 N. B. It need not be observed, that the English translations heri 
 larly the case in the passive imperfects, for which we possess no ph 
 imperfect by converting the ablative of the agent into the nominativ 
 
 The passive and middle voices are here taken as the same in sign 
 with a separate form, because the pronominal termination, which mr 
 
SCHEME OF TENSES. 
 
 PRESENT TIME. 
 
 PAST TIME. 
 
 r Absolutely. 
 
 flMPERFECT.< 
 
 (^ Relatively. 
 
 
 Inceptive. 
 
 ^Perfect. 
 
 r Absolutely. 
 Imperfect. < 
 
 C_ Relatively. 
 
 Inceptive. 
 
 n 
 
 Perfect. 
 
 Active. 
 
 J rvvTio, verbero, 
 i^ I am beating. 
 5 TBTvtpa, verberavi, 
 ( I am having beaten. 
 
 Tv-lpo}, verberabo, 
 I am going to beat. 
 
 S rvTTTa), verberem, 
 ^ I may be beating. 
 5 TiTv^u), verberaverim, 
 ^ I may be having beaten. 
 
 S Tvxpeia, verberaturus sim, 
 \ I may be going to beat. 
 
 ^ Tvtpo), verberaverim, 
 \ I may finish beating. 
 
 Passive. 
 
 ^ TVTTTOfxai, verberor, 
 I I am being beaten. 
 S Tsrviifiai, verberatus sum, 
 ( I am in a beaten state. 
 
 CTvypofiai, or, rv^Brjaofiai, 
 
 ) verberabor, or, verberatus ero, 
 
 "^ I am going to undergo a beating, or, 
 
 V to be in a beaten state. 
 
 S riiTrrwjuai, verberer, 
 
 i^ I may be being beaten. 
 
 ^ TiTVf.ifX£V0Q w, verberatus sim, 
 
 \ J may be in a beaten state. 
 
 C TVxpbJfJLai, or, 
 
 ? rvf6a>, verberatus sim, 
 
 (^ I may finish being beaten. 
 
 Active. 
 
 < tTvvTov, verberabam, 
 I I was beating. 
 S Brtrixpeiv, verberaveram, 
 ^ I was having beaten. 
 
 tTv\pa, verberavi, 
 T beat or did beat. 
 
 TVTTToifit, verberarem, 
 
 I might be beating, 
 I TeTV(poifxi, verberavissem, 
 [ I might be having beaten. 
 
 ^ Tvipoifii, verberaturus essem, 
 \ I might be going to beat. 
 
 TVil/ai/jii, 
 
 I might finish beating. 
 
 ^ krvTTTOfirfv, verberabar, 
 i^ I was being beaten. 
 S irtTvi.int]v, verberatus eram, 
 ^ I was in a beaten state. 
 
 ^ Irv^Gjjv, verberatus eram, 
 I I was beaten. 
 
 i TVTTroifirjv, verberarer, 
 
 ( I might be being beaten. 
 
 S TETviifiivoi; hrjv, verberatus essem, 
 
 i^ I might be in a beaten state. 
 
 I 
 
 Tv^o'inriv, or, TV(p9t]aoi(iriv, 
 
 verberandus essem, 
 I might be going to suffer a beating, 
 
 or to be in a beaten state. 
 
 C TVipaifirjv, or, 
 
 ? TV^Oeirjv, verberatus essem, 
 
 (_ I might finish being beaten. 
 
 N. B. It need not be observed, that the English translations here are not such as can be legitimately employed with elegance. The precise sense is all that is aimed at. This is particu- 
 larly the case in the passive imperfects, for which we possess no phrases. " I am or was being beaten" is neither grammatical nor is it sense. And for it we must always substitute an active 
 imperfect by converting the ablative of the agent into the nominative, " They were or are beating me." 
 
 The passive and middle voices are here taken as the same in signification, and frequently convertible. The first aotist seems to be the only tense essentially reflective, and to be provided 
 with a separate form, because the pronominal termination, which makes it reflective, does not enter so clearly into the first aorist passive as into the other tenses of that voice. 
 
SCHEME OF TENSES. 105 
 
 the perfect with the genitive. And, as before 
 observed, it becomes the more probable when 
 we find that in languages which possess cases, 
 these tenses are also found, and that in others 
 they are both wanting together. 
 
 It is needless to illustrate the identity of 
 those mental operations by which we receive 
 perceptions in the order of time and the order 
 of space. In both cases our states of mind 
 follow one another as the links in a chain, and 
 are susceptible of the same modifications and 
 phases. A notion of time may indeed exist 
 without that of space or extension, but the latter, 
 in the present constitution of our muscular sys- 
 tem, cannot be obtained without the former. 
 In themselves, however, they are generated by 
 successive afi'ections of the mind, and involve 
 no essential diff'erence. How it is that there 
 are but three inflections for the cases, and four 
 for the tenses, is evident ; since the subdivision 
 of those tenses, which imply the actual exist- 
 ence of qualities as either being or superinduced, 
 has no counterpart in the case of substances. 
 
 In applying these observations to the eluci- 
 dation of the Greek tenses it must be remem- 
 bered, that the retrospective, existing, and 
 the prospective connection of a quality with an 
 object, necessarily involve some notion of time. 
 The past, the present, aad the future, are essen- 
 tially comprised in the abstract perfectness, im- 
 
106 SCHEME OF TENSES. 
 
 perfectness, or inceptiveness of a fact. — And 
 the nature of things in particular instances will 
 prevent the formation of a full and perfect sys- 
 tem of the four tenses, described under each 
 head of present and past time. But if we can 
 procure any modification of the verb which ex- 
 presses a fact with reference only to one 
 time, or rather to no time at all ; and if in this 
 mood such a system is fully developed, we 
 may assume it as a legitimate test of the cor- 
 rectness of the hypothesis. Such a mood is 
 found in the infinitive. Whatever be its origin 
 its use is obvious. Conjecture might perhaps 
 suggest its formation from the neuter of the 
 participle — much in the same manner as we 
 form our English expressions, the doing, the 
 striking, the having fought. And the other 
 form of our infinitive, consisting of the verb with 
 the particle to prefixed, seems rather analogous 
 to the construction of the Latin infinitive from 
 the verb of motion eo, than to be, as some sup- 
 pose, a construction with the Greek article in 
 its original state A mere fancy, however, of 
 this kind deserves but little attention ; al- 
 though it is certainly curious that the Latin 
 language possesses, like the English, two 
 classes of infinitives — the Sanscrit supines, 
 and its own termination in re, and isse, the 
 respective formation of which seems perfectly 
 to accord with the origin of ours. 
 
SCHEME OF TENSES. 107 
 
 However this may be, the infinitive is 
 evidently a noun, and a noun substantive. It 
 is used w^hen attention is called not so much 
 to the attribute itself, as to its formation or ex- 
 istence. And its very abstract character seems 
 to be the cause of its excluding any generic 
 termination, and consequently being iucapable 
 of inflection into cases. As a noun it cannot 
 essentially involve any notion of time ; more 
 than the idea of a book, a horse, or a meadow. 
 But it expresses the state of the quality under 
 its four heads, of completion, of incipiency, 
 and of existence — and of existence either abso- 
 lutely in itself, or relatively to some superin- 
 ducing cause ; Tt;^^;, to finish heating, a perfect 
 action ; Tvi>eiv, to be going to beat, an inceptive 
 a.ction ; r^Vrefv, to beat, an action present, and con- 
 sequently imperfect ; rerocpey^i, to ixtain the attri- 
 bute of beating previously assumed, and conse- 
 quently to be likewise in an imperfect state. 
 What the infinitive expresses in an abstract 
 form the participle expresses in the concrete, 
 and under the same four heads. — And here also 
 it is evident that the notion of time is merely 
 accidental, since otherwise there would be se- 
 parate inflections for participles of the past, 
 and participles of the present time, such as we 
 shall find to exist in the indicative and sub- 
 junctive moods. The diflference between verbal 
 adjectives and participles, evidently lies in the 
 
108 SCHEME OF TENSES. 
 
 significations here attributed to the inflections 
 of the tenses. 
 
 The next mood in which we may clearly as- 
 certain the meaning of these tenses is the 
 imperative. It is here that many philologists 
 seek for the root of the verb ; and although it 
 is perhaps more correct to consider its several 
 shapes as parallel than as derivative formations, 
 it is certain that here we shall naturally find 
 the root of the word in its most compact and 
 abbreviated form. The expression of a com- 
 mand like that of a want, is naturally the men- 
 tion of the thing wanted — and nothing more. — 
 And since to command a thing which is either 
 past or present is an absurdity, the impera- 
 tive, if time be looked to, must universally re- 
 late to the future. Instead, however, of any 
 future signification being annexed to the tenses 
 in this mood, we find that the only one omitted 
 out of the four to be anticipated, is that which 
 is usually termed the future. There are but 
 three inflections, rvitre, r^ov, and reVu^e, be beat- 
 ing, have beaten, or jinish beating, and con- 
 tinue having beaten, if such a rough transla- 
 tion may be allowed. That the inceptive or 
 imperative of T^4/w should be omitted is per- 
 fectly natural, since we never desire that an 
 object which we want should be removed from 
 us by any interval. If we wished to see a 
 person painting a picture, we should say ypa^e. 
 
SCHEME OF TENSES. 109 
 
 If we wished the picture finished, and the at- 
 tention of the painter directed to some other 
 object, we should say ypa^pov. And if we wished 
 to see him precisely at the moment when the 
 colours were wet, and his brush just laid down, 
 we should say yeypa^e. When we proceed to 
 follow up this system into the indicative mood, 
 we find it for the first time branch out into two 
 heads : and the notion of time immediately an- 
 nexed to it in the shape of the augment. 
 
 And we also find that the scheme for 
 present and the scheme for past time, are 
 each defective in one tense. Instead of eight 
 tenses, as we should perhaps anticipate, there 
 are but six in the indicative. 
 
 Peesent. Past. 
 
 Perfect. ■ ■ "Ervypa. 
 
 Inceptive. Tv^o). 
 
 ^ C Absolutely. TwTrrtD. "Etvtttov. 
 
 Imperfect, < ^ _, , , 
 
 i Kelatively. TsTvtpa. ETfTU(peiv. 
 
 The deficiency of the perfect tense in the 
 present is obviously necessary, since the very 
 notion of completion implies past time. — ^And 
 the defect of the inceptive tense in the past, 
 may also be accounted for by remembering 
 how little we dwell upon past anticipations un- 
 fulfilled. — The tendency of an object to a par- 
 ticular state at the present time makes a great 
 impression on our minds. But afterwards it 
 soon escapes us. 
 
110 SCHEME OF TENSES. 
 
 The several significations of these tenses, 
 deducible from their primary meaning, are too 
 well known to require much illustration. The 
 inceptive admits of very few deflections. Of the 
 two present imperfect tenses, rr^irrco and rervipa, 
 the former is employed to signify not only the 
 connection of two qualities at the present mo- 
 ment, but from hence their constant connection 
 at any imaginable moment. Hence the use of 
 the present in necessary and identical proposi- 
 tions.^ — It also signifies continued existence, 
 frequent repetition, and sometimes an ineffec- 
 tual effort. The reduplication in the present 
 perfect as rert^a, seems analogous to that of the 
 superlative degree in adjectives ; and to denote 
 the continuance of the effect of a past action up 
 to the present time. Since as excess in degree 
 is denoted by the repetition of the primary 
 idea, so any continuity either of duration or 
 extension is perceived in the same manner. 
 The termination in k and a naturally connects 
 itself with the formation of our own perfect 
 tense by the verb have, and probably was de- 
 rived from a similar origin. The notion of 
 possession implying previous acquirement ; and 
 these two ideas being precisely the significa- 
 tion of the perfect tense. 
 
 Of the past tenses, the first aorist signifies in 
 the first place merely a previous connection of 
 the attribute with the subject. — And since such 
 
SCHEME OF TENSES. Ill 
 
 instances in past time are the basis on which 
 we reason to general coincidences, it implies 
 custom and habit. And the past imperfect 
 tenses are analogous in their uses to the present 
 imperfect. 
 
 There seems no reason why in this scheme 
 of tenses any distinction should be made be- 
 tween the aorist, and what some persons have 
 termed the oristic tenses. We may mark out 
 the place of an object either by reference to 
 ourselves or to some other fixed point; or more 
 precisely to both ; the book here, the book on 
 the table, or the book here on the table. 
 What our own person is in the order of space, 
 the present moment is in the order of time. 
 We may describe a fact as present or past, or 
 we may add the precise fact with which it was 
 coincident, or concomitant. — Sometimes this 
 fact is expressed, sometimes implied in the 
 context. But its insertion does not appear 
 essentially necessary to any one tense more 
 than to another. 
 
 In this sheme it is supposed that the second 
 aorist and second future have no signification 
 different from the first forms of those tenses. 
 Certainly none is perceivable, and it would 
 perhaps be advisable to examine whether or 
 not the other tenses are not severally to be 
 considered as so many distinct roots, and not 
 derivative inflections from each other. The re- 
 
112 SUBJUNCTIVE MOOD. 
 
 duplication of the paulo post futurum in the 
 passive voice, seems to mean the same as the 
 reduplication in the preterperfect and second 
 aorist, viz. the continuance of the effect. And 
 its absence in the active voice is perfectly na- 
 tural, since no such continuance would be visi- 
 ble in the future conduct of an agent. 
 
 It only remains to consider the nature of 
 those forms v^hich are denominated the opta- 
 tive and subjunctive moods. To analyze the 
 subject fully, and illustrate it with examples, 
 would very far exceed the plan of the present 
 sketch. — All that can be done is to propose a 
 few conjectures, which the student himself 
 must confirm or refute. 
 
 And first, as we have pursued in other cases 
 the analogy between the noun and the verb, 
 we may recur to it here also. — In English 
 there are no cases and no tenses ; in Greek 
 there are both. It was probable that a similar 
 notion was in this instance expressed or 
 omitted. In English there is no distinct in- 
 flection to represent the second of two 
 nouns ; and no distinct inflection to serve for a 
 subjunctive mood. In the Greek there are. 
 It is therefore not impossible that as the ob- 
 lique case stands to the nominative, the sub- 
 junctive mood may stand to the indicative ; 
 that it may represent the second of two facts, 
 as the oblique case represents the second of 
 
SUBJUNCTIVE MOOD. 113 
 
 two objects : and that as in the one instance 
 the two objects are closely connected to the 
 senses ; so in the other the two facts may go 
 hand in hand as cause and effect. The similar 
 nature of the inflection in both deserves to be 
 noticed. Supposing then that the absurd no- 
 minal distinction between the optative and sub- 
 junctive may be removed, we shall find that 
 when thrown together they furnish us with 
 eight tenses. And these may at a glance 
 be divided according to the scheme before laid 
 down, so as to supply a complete system of 
 perfect, imperfect, and inceptive forms to each 
 of the two divisions of time, the past and the 
 present. 
 
 Past. Present. 
 
 Perfect fTv^ai[jii. fTiJi/zw. 
 
 iNCErTIVE 
 
 Imperfect 
 
 iNCErTivE J Tj:i4/oijwt. J Ti^eia. 
 
 I ^ TVTTTOIIJI. I ^ Til 
 
 L t Tervcpoifii. \^ \ Tet 
 
 Or, in English, he j^Finisli striking.^ fFinish striking. 
 
 went that he 
 might 
 
 Be about to strike, J Be about to strike. 
 
 Be striking, ^ i Be striking. 
 
 t,Have just struck. • l^Have just struck. 
 
 And it may be worth observing, that the 
 omission of the augment in this mood is favour- 
 able to a previous hypothesis. Since the date 
 of this second fact, if dependent on a former 
 one, must be in all cases future to it, and con- 
 sequently cannot be definitely marked. 
 
 It is evident that to* confirm this hypothesis. 
 
114 SUBJUNCTIVE MOOD. 
 
 would require voluminous illustrations. — And 
 all that can be done must be left to the obser- 
 vation of the student. With respect to the 
 aoristus ^olicus which I have ventured to 
 rank as the subjunctive of the inceptive, or 
 future, such a conjecture appears probable 
 from its form, and as far as the confined read- 
 ing of a single individual has extended, it is 
 borne out in every instance, and in many is ab- 
 solutely necessary to preserve the sense. A 
 few observations may be made respecting the 
 use of the subjunctive in itself. 
 
 First then, the former fact on which the se- 
 cond expressed in the subjunctive is supposed 
 to depend, is frequently omitted. It is omitted 
 whenever it is not known. — Just as we find the 
 genitive case employed to indicate the part of 
 a whole, the part itself not being expressed 
 from its being unknown. The verb to happen 
 may here be introduced, and give the full force 
 of the subjunctive. 
 
 Secondly, whenever a number of individuals 
 are capable of the same predicate, and the 
 choice is to fall upon one only, the verb will be 
 put in the subjunctive, because the choice 
 must be determined by some circumstance not 
 yet known. 
 
 Thirdly, whenever a case is conceived likely 
 to occur, the verb of the subjunctive will be put 
 in a present tense — but if unlikely, or a fact 
 
SUBJUNCTIVE MOOD. 115 
 
 which is known not to exist, the verb will ap- 
 pear in a past tense. — The hypothesis will be 
 formed apart from any existing circumstances, 
 and the past time being the only one distinct 
 from the present, and marked by separate in- 
 flections, a case not existing will naturally be 
 thrown under it. The same principle acts in 
 English^ — and it accounts, in both languages, 
 for the expression of wishes not likely to be 
 gratified, in the past tenses of the subjunctive. 
 The common translation of the tenses here 
 considered as past, by the signs ivould, could, 
 and should, which are themselves the past 
 tenses of ivill, can, and shall, evinces the 
 analogy. 
 
 Fourthly, these wishes will themselves be 
 expressed in the subjunctive, since they are 
 facts which are conceived to depend upon the 
 will or exertions of another. 
 
 Fifthly, as a past fact, A, may produce ano- 
 ther past fact, B, and also a present fact, C ; 
 but a present fact, D, can only produce a pre- 
 sent fact, E, (for the future has no inflection), we 
 may see the reason why and how far the prin- 
 ciple laid down by Dawes is correct; that a 
 past tense in the first clause requires an opta- 
 tive in the second ; and a present in the first, a 
 subjunctive in the second. The real fact is, 
 that a past tense may precede both an optative 
 and a subjunctive, according to circumstances. 
 
 I 2 
 
116 SUBJUNCTIVE MOOD. 
 
 But a present tense can only precede a sub- 
 junctive, unless another nonexisting hypothesis 
 is understood on which to found an optative, or 
 some idiom of the language accounts for the 
 exception. Let L be the middle letter of the 
 alphabet, and stand for the present moment, 
 dividing time into two portions, E may be the 
 cause of K and also of R, but N can only cause 
 S or T on that side of L. 
 
 Sixthly, it is very necessary to fix on the 
 precise fact on which the second is built, and 
 to mark the time which is taken as the stand- 
 ard, particularly with reference to what gram- 
 marians term the oratio obliqua, so common in 
 the Greek idiom. 
 
 And seventhly, it may be worth while nicely 
 to distinguish those cases in which the indica- 
 tive is admitted into the second clause instead 
 of the subjunctive; and to observe how fre- 
 quently this takes place in the case of negatives 
 which cannot act, and in apparent effects, which 
 in reality are not effects at all, according to 
 the notion of antecedent and consequent esta- 
 blished among the Greeks. 
 
 The whole nature, however, of the subjunc- 
 tive mood, might well deserve a separate dis- 
 cussion. And the present notices can throw 
 but very little light on its complicated and im- 
 portant uses. 
 
 After the analysis of the verb into the attri- 
 
GREEK PARTICLES. II7 
 
 bute and pronoun, it is unnecessary to explain 
 the nature of the middle voice, or its various 
 significations. They may all be comprised in 
 the fact that the pronoun is sometimes taken 
 for the accusative, and sometimes for the 
 dative. And the frequent intermixture of 
 tenses which have wrongly been separated 
 from each other under the two distinct heads 
 of active and passive, is easily to be explained 
 in this manner. It would be very desirable 
 to account for the deficiencies in the tenses of 
 particular verbs by looking to their intrinsic 
 nature. Those in the verb eT/>oi are singular, and 
 deserve attention. And in general they are to 
 be traced not to mere accidental omissions or 
 usages, but to the nature of things and the 
 principles of the human mind. 
 
 It only remains for us to take a cursory view 
 of those little words or particles which enter so 
 largely into the Greek language; and which 
 to many appear both insignificant and useless. 
 Even these, however, petty and unmeaning as 
 they seemingly are at present, we may be 
 assured possessed originally some certain and 
 positive signification. They must have been 
 ranked under one of those classes which have 
 before been analyzed. What their etymological 
 origin was it is almost impossible to ascertain, 
 both from the abstractedness of their meaning 
 and the simplicity of their form. And little 
 
118 GREEK PARTICLES. 
 
 more can be done than to throw them into 
 something like order ; and form some probable 
 hypothesis on their primary senses, and subse- 
 quent deflections. 
 
 If we look then to the operations of the 
 human understanding, they may easily be re- 
 duced under the following heads. For the 
 mind is either sensible of single distinct per- 
 ceptions of pleasure or pain, which give birth 
 to interjections, as involuntary and irrational 
 sounds : or it receives from the senses a num- 
 ber of perceptions which it groups together, 
 and anticipates by the law of association, ac- 
 cordingly as they coexist in space or are con- 
 secutive in time. 
 
 And it need not be observed, that the ex- 
 pression in language of any judgment, or any 
 reasoning, must imply the previous formation 
 of such groups of ideas in the mind. The 
 particles then, if a definition be required, may 
 be considered as words drawn from some other 
 use, and employed analogically to denote 
 certain accidental results of these mental ope- 
 rations. They do not express in their second 
 office any distinct ideas, but are signs of certain 
 states of mind which occur when it is repeating 
 any trains of ideas already associated by expe- 
 rience. And they must originally have stood 
 for tangible and visible objects, and been trans- 
 ferred from thence by some analogy or another. 
 
CONNECTIVE PARTICLES. 119 
 
 since A could scarcely explain to B any in- 
 ternal perceptions of his own mind, except by 
 placing before B the same external objects 
 which would act upon others in the same man- 
 ner as upon himself. 
 
 First then, we may express in words either a 
 group of ideas obtained by that process which 
 the logicians term simple complex apprehen- 
 sion ; or the result of a comparison between 
 two groups or series of perceptions. And in 
 the former case it would seem, that when the 
 mind in forming a complex substaiitive has 
 previously passed uninterruptedly from one 
 state into another, till all the links in the chain 
 were run out, it acquires a tendency to pass 
 successively from one into another; just as we 
 expect, anticipate, and are ready to fall into 
 the notes of a well-known tune before they are 
 played, are disappointed if it suddenly breaks 
 off, and feel no farther tendency of the kind 
 when it comes to its natural close. This 
 momentum, as it were, which the mind ac- 
 quires, seems in Greek to be expressed by the 
 word Koi. Of its etymology it would be absurd 
 for any one but a professed linguist to assert 
 anything. The word qucero in Latin might 
 seem from its use to be derived from Ka\, and an 
 obsolete verb of motion connected with eo, ire, 
 and to signify a constant progressive advance. 
 But an hypothesis is not worth supporting 
 
120 CONNECTIVE PARTICLES. 
 
 which rests on such a slender reed. And we 
 can only say with certainty, that if we could 
 affix to the word with propriety the sense of 
 go 071, or advance, it would explain all the 
 uses of this important particle. Cicero was 
 eloquent, Ka\, a patriot, ku), a philosopher. As if 
 the hearer was told to go on, not to consider 
 the series of accumulated qualities to be yet 
 exhausted. From hence it might naturally 
 signify also, not only B, but, go on, something 
 beyond this, C also. In the same manner it 
 would stand for even. He was cruel not only 
 to strangers, but, go on, something farther and 
 beyond, even to his own children. So likewise 
 it signifies immediately, like the Latin et and 
 atque. This was done, go on, without any 
 interruption, something else happened. So too 
 its sense of although; Demosthenes /caircp an 
 orator was not courageous — go on, allow, do 
 not hesitate, as you may be inclined to do. 
 And the English notwithstanding, expresses pre- 
 cisely the same notion, namely, the absence of 
 an anticipated obstacle to a farther advance* 
 Hence, too, it is employed to mark the conces- 
 sion of a point, which was capable of dispute ; 
 and may be translated into English by an em- 
 phasis on the verb — if it he so, e* koX Un. And 
 the same radical notion runs into all its uses ; 
 as, for instance, in the expression of similarity ; 
 its connection of qualities combined in the 
 
CONNECTIVE PARTICLES. 121 
 
 same substantive ; of substantives only nume- 
 rically distinguished from each other ; and 
 sometimes v^hen prefixed to the apodosis of a 
 sentence, especially in the idiom of the Septua- 
 gint ; in all of which cases there is an easy 
 transition of the mind through the several ideas 
 specified, from the very first principles of the 
 law of association. 
 
 Besides, however, the conjunctive particle 
 Kal, the Greeks possess another, re. And the 
 peculiarities in its use well deserve a minute 
 examination. — 
 
 First, it is singular in its position, as sub- 
 joined, not prefixed, to the noun. 
 
 Secondly, it is curious that when found in 
 combinations with kcu, it invariably occupies the 
 first place, and never the second. 
 
 Thirdly, it is connected in a very remarkable 
 manner with the relative pronoun, where it 
 appears to be, what it never assuredly could 
 be, otiose and superfluous. 
 
 Fourthly, its use as a conjunctive, even when 
 coupled with other disjunctive particles, is very 
 difficult to explain. 
 
 And fifthly, there seems to prevail in it a 
 marked distinction from kcc), in its coupling two 
 equivalent terms, where ku) implies an excess 
 in each progressive step. 
 
 Now there can ba little doubt that as the 
 Latin et, probably from the same root as eV*, 
 
122 CONNECTIVE PARTICLES. 
 
 corresponds with koc), so the other conjunction 
 of that language, que, is the counterpart of re. 
 But the analogy is much more striking in the 
 second case than in the first. 
 
 And if we look to the peculiarities in the use 
 of xe, and to the every-day operations of the 
 mind, we may, perhaps venture an hypothesis 
 as to its real nature. Supposing then a person 
 was relating a certain number of facts to an 
 impatient auditor, or at any rate to one whose 
 attention was likely to terminate at each suc- 
 cessive link, what would the speaker naturally 
 do? what is it that we all do in such a case? 
 — Simply this — the moment that we come to 
 the close of one fact, we instantly subjoin the 
 words something else : we add to it a sign, 
 which, without explaining what is to come, 
 states that something is coming; that some- 
 thing being known and determinate to the 
 speaker; but not rising in importance above 
 the preceding point, or likely, when enun- 
 ciated, to cause any hesitation in the hearer. 
 
 Now it is well known that the old Greek 
 pronoun of the third person was e, and the more, 
 modern form sufficiently accounts for its be- 
 coming Te. It is also evident that the k derived 
 from the old Greek, is retained in the Latin 
 definite pronoun quis, or quidam, while the t 
 appears in the Greek t*^. And que is to quis, 
 as T€ to TK. And the absence of any such con- 
 
CONNECTIVE PARTICLES. 123 
 
 junction in our own language may be easily ac- 
 counted for by our want of any short prono- 
 minal substantive of the same kind. Although 
 our definite article the is probably the same 
 word deprived of its gender. 
 
 If we were to apply this hypothesis, we 
 should probably find it exactly resolve the 
 phenomena to be accounted for. 
 
 Let C and D be two terms to be coupled. 
 To C we subjoin re, to signify that something 
 else is coming; and if D is an equivalent to C, 
 if there be no hesitation anticipated in the 
 mind of the speaker when passing from one 
 to the other, D is then added, with i-e affixed 
 to it likewise, to show that it is the something 
 alluded to, exactly in the same manner as the 
 relative is employed both in modern and 
 ancient languages. But if D is an increase or 
 advance upon the former step, ku) is introduced 
 and re omitted as superfluous ; since, by the 
 insertion of koi, D is sufficiently marked out as 
 the prospective point intended by re. The dis- 
 tinction is marked in Latin by the analogous 
 forms turn, and turn, cum, and turn. 
 
 In conjunction with the relative pronouns, 
 with eVe;, iq, o7o(;, el, aud othcr words of pronominal 
 form, it seems to have precisely the same use, 
 position, and nature, as the indefinite r*? in oVt*^, 
 the neuter of which it i^ assumed to be. 
 
 Sometimes it appears to connect a second 
 
124 CONNECTIVE PARTICLES. 
 
 clause with a former, by throwing the mind 
 back to the former. And sometimes to be sub- 
 joined to the second clause instead of appearing 
 in both, when the second is merely an accessary, 
 which we are not anxious to connect with the 
 first ; so that in this latter case re answers the 
 simple purpose of the connective and, while in 
 the former, when we wish to fix the mind 
 not merely, on the separate existence of two 
 qualities in the same subject, but their coin- 
 cidence, and this particularly where they are 
 not often found together, re answers the pur- 
 pose of both in English. 
 
 The full development of all its uses would, 
 however, require a long and elaborate in- 
 quiry. The general principle may, perhaps, be 
 summed up in this. That whenever any hesita- 
 tion is anticipated in passing on from one term 
 of a series to another, koi is employed. But when 
 the terms are merely to be connected, they 
 are simply placed in juxtaposition to each 
 other ; each successive link being anticipated 
 before it is accurately marked out, by means of 
 the indefinite pronoun attached to the pre- 
 ceding link. 
 
 When, instead of running either groups of 
 qualities, or similar substances and facts to- 
 gether, we wish for any purpose to separate 
 them one from the other, the particle ? is em- 
 ployed. And in its use it is very analogous to the 
 
NEGATIVES. 125 
 
 English or, which might, perhaps, be connected 
 with the root of the Greek o>o?, a line or boun- 
 dary, separating two things. Sometimes then 
 ^ will be introduced where each member of an 
 enumeration is to be separately dwelt upon, 
 either for the purpose of amplification or clear- 
 ness. Sometimes it is prefixed to a second 
 case merely to mark it as distinct from one 
 which preceded. Sometimes when a choice is 
 offered it will separate the alternative abandoned 
 from the one selected. And still more fre- 
 quently where there are, as it were, several 
 candidates for a single object which can fall 
 only upon one, it will mark the perfect incom- 
 patibility of the several possible cases. It need 
 only be observed, that in instances of compari- 
 son, the sign of the comparative degree is per- 
 fectly immaterial ; and that, introduced, it in 
 no way is capable of throwing light upon the 
 construction. 
 
 It is self evident that before any notions can 
 be connected or disjoined, they must be per- 
 ceived to be distinct one from the other. And 
 in placing the particles which mark these 
 mental operations before those which are em- 
 ployed in making distinction, a false arrange- 
 ment has perhaps been adopted. This, how- 
 ever, is of no great importance. 
 
 Of distinctive particles there appear to be 
 two kinds, those which are employed in the 
 
126 NEGATIVES. 
 
 office of making distinctions, and those which 
 mark them when made. 
 
 It has before been observed that the mode 
 by which we become sensible of a difference 
 in two objects, is by finding some break in the 
 chain of our anticipations. If a series of six ideas 
 has been obtained from one substance, and only 
 five occur in another, we perceive that they are 
 difi'erent. And it is quite evident that this 
 alteration may take place in two ways, either 
 by a deficiency in our second series, or by the 
 intrusion of an extraneous link. We may 
 either anticipate a perception and find a blank, 
 or anticipate a blank and find a perception. 
 
 This appears to be the primary difference 
 between 'Iv and i^ri — and all the complicated uses 
 of these words seem to be explicable by a little 
 attention, if this principle is steadily kept in 
 view. — Hence it is that oy is employed in cate- 
 gorical propositions, to indicate the absence of 
 a quality which had been suggested to exist in 
 the subject, but i^i in imperative forms, in 
 expressions of wishing, forbidding, and depre- 
 cating. Hence also /xv? is used with adjectives, 
 with imperative moods, and with subordinate 
 clauses — and is generally to be construed by 
 the English but, or without. And when the 
 two are found together, the distinct meaning of 
 each is usually to be retained, except perhaps 
 in some few cases which have perplexed critics; 
 
DISTINCTIVE PARTICLES. 127 
 
 and in which the ov supposed to be superfluous 
 is introduced, because the end of the action or 
 subject of consideration is not merely privation, 
 but negation founded on privation. The dis- 
 tinction is difficult and abstruse without ex- 
 amples. — But it may easily be observed by 
 referring to the familiar instances accumu- 
 lated in grammatical works ; and it will assist 
 the search if we examine each instance, and 
 inquire how far the fact, to which the two ne- 
 gations are applied, is considered as an active 
 cause or any thing positively existing. The 
 negation of colour n^ay act still as another 
 colour, but the mere detraction of it can do 
 nothing. And this seems to be the clue to the 
 whole mystery. The origin of the words /xtj and 
 ov is of no very great importance. Mtj is per- 
 haps to be found in the root of the verb from 
 which lA'fiv, fA,du, and in Latin 77ioveo are derived. 
 And ov is probably a mere interjection. — Even 
 trifles are sometimes worth attention, and the 
 natural movements of the head in the expres- 
 sion of negation or affirmation, according pre- 
 cisely as they do with the gestures with which 
 w^e beat time to a rhythmical tune, or express 
 our pain at a false note, were perhaps the first 
 cause of those sounds which respectively indi- 
 cate assent and dissent. 
 
 When distinction has thus been attained by 
 means of negation, the next thing is to mark 
 
128 DISTINCTIVE PARTICLES. 
 
 it, and to fix the attention of a hearer upon 
 each point separately. This is of course most 
 necessary when the two terms from their simi- 
 larity are most likely to be confounded to- 
 gether. And hence we generally find the dis- 
 tinctive particles applied to individuals of the 
 same species, when they differ in some point 
 in which it was likely for them to agree. — ■ 
 Sometimes also they are employed when ap- 
 parently there is but one term ; and the atten- 
 tion is to be fixed upon that, without running 
 on to others of the same class. ^ — Now as in 
 forming our notion of number we necessarily 
 perform this process of discriminating between 
 individuals very similar, and of keeping them 
 apart in our minds, analogy would naturally 
 suggest the symbols of number as marks of 
 distinction ; and what in English is expressed 
 by the phrases^r*^ and second, the Greek seem 
 to have expressed by one, two, /^ev, and u>, Of 
 these iwev is evidently the neuter of I*?; and U the 
 counterpart oitoo, seems to bear the same rela- 
 tion to ^vo, as re to to'. Some observations pre- 
 viously made on our perception of number will 
 show why the Greeks did not distinguish be- 
 yond two- members of a class. ^ — The very nature 
 of the words explains the mode in which U be- 
 comes at once a connective and a distinctive 
 particle — and all their various uses will readily 
 appear when these significations are affixed to 
 
ILLATIVE PARTICLES. 129 
 
 them. To enumerate them would be not so 
 much to explain the employment of the par- 
 ticles, as to mention all the occasions in which 
 the mind wishes to dwell separately upon two 
 objects. 
 
 The particle S,v which recurs so frequently in 
 composition, expresses a still greater difference 
 than U. — And its various senses' of back, again^ 
 contrary to, and opposite to, are fixed by the 
 same analogy which has combined them under 
 the same words in other languages. 
 
 The origin of axxa is self evident, and requires 
 no illustration. But it may be worth while to 
 observe here, that the connective particle is 
 acknowledged to be a pronoun, and coupling 
 this instance with many others of the same 
 kind, to infer the probability of this process in 
 other cases where the derivation of the word is 
 not so palpable. 
 
 Illative particles are the last class to be here 
 mentioned. When the operation of reasoning 
 is stripped of all its mystery, it is nothing but 
 the anticipation of a second fact from a former ; 
 this anticipation being caused by previous ex- 
 perience of such a conjunction; and being re- , 
 gulated by the law of association. — An unin- 
 terrupted experience wiir rivet, as it were, the 
 two facts together, so that they will occur in 
 unbroken succession to the mind, and will be 
 expressed consecutively in the same manner. 
 
 K 
 
130 ILLATIVE PARTICLES. 
 
 But an experience which has been interrupted 
 will keep the reason fluctuating, as it were, be- 
 tween the two — and in a certain degree sus- 
 pend the anticipation. In this case our ten- 
 dency to infer and to believe a fact is regulated 
 by two simple laws. We are inclined to believe 
 that which is agreeable to our feelings, and that 
 which coincides with our preconceived trains of 
 association. In the former case a fact may be 
 called probable, in the latter likely — meaning 
 by probable, that fact which meets our ap- 
 probation — and by likely, that which is like, 
 and resembles others. The two words are 
 frequently confounded. But it is of the highest 
 importance to keep the two principles distinct. 
 This likelihood and probability appear to be 
 expressed in Greek by the particle a,v — and if 
 etymology could trace it up to any connection 
 with the root of the verb a,vUv(o to please, it 
 would present a very curious, but very natural 
 coincidence with our own words likely and 
 agreeable, which express both meanings at 
 once. The degree of likelihood is marked in 
 Greek by adverbs, as tVo-?, and rd^a; the equal 
 balance of the mind, when suspended between 
 two opposite experiences, being perhaps de- 
 noted by the former, and its rapid tendency to 
 form an inference by the latter. A still farther 
 degree of certainty seems to be contained in 
 a(>a, probably from ap apto. In analytical rea- 
 
ILLATIVE PARTICLES. 131 
 
 soning where the conclusion is stated or im- 
 plied first, and the premiss subjoined, the same 
 apa combined with ye, is used when the inference 
 is conceived to be valid; and ye alone is an- 
 nexed to the premiss, where it is a particular 
 fact leading to a general conclusion. As the 
 process of production in nature from an embryo 
 to a mature creation, is precisely analogous to 
 the act of generalization from a single fact, a 
 very fanciful etymology might perhaps attempt 
 to connect the particle ye with the root of the 
 Greek verb of production. But this is too slight 
 a foundation to rest on. 
 
 With respect to synthetical reasoning, both 
 a view of the mental operations of which it con- 
 sists, and the analogy of other languages will 
 show, that there are no particles primarily 
 formed to express inference. The fact on 
 which the conclusion is built may be, and 
 generally is, repeated in the shape of the pro- 
 noun — and this in various cases. In the da- 
 tive, when the second fact is included in the 
 former ; in the genitive, when the former is the 
 efficient and active cause of the second ; in the 
 accusative, when it is merely the occasional 
 cause — and the two particles Iw and h^i are 
 sometimes introduced, just as the English pro- 
 nouns of times, then and now ; the former di- 
 recting the attention to the premiss, the latter 
 to the conclusion. Whatever shapes these 
 
 . K 2 
 
132 ILLATIVE PARTICLES. 
 
 particles or any other words may assume, we 
 may be assured that there is one primary mean- 
 ing running severally through all. To trace this 
 out, and reduce them into their simple and pri- 
 mary element, is an exercise for the mind of all 
 others most fitted to develop its faculties, and 
 improve its habits of thinking. It is the prin- 
 cipal purpose for which dead languages are 
 studied, that we may use them as a field in 
 which to sharpen and construct the instruments, 
 which are afterwards to be employed in other 
 more practical occupations. And it is with this 
 view, and in the hope of suggesting materials for 
 thinking, and hypotheses for inquiry, rather than 
 from a presumptuous confidence in what has 
 been advanced, that the preceding conjectures 
 have been ofi*ered. — And if any student should 
 be induced to undertake for himself the exami- 
 nation of their correctness, the purpose of this 
 little Essay will be sufficiently answered, 
 whether they are confirmed, or refuted. 
 
APPENDIX. 
 
 (A). The existence at present of a language addressed to 
 the eye and yet sufficiently abstract to avoid these inconveni- 
 encesj does not controvert this supposition. — It is difficult, in- 
 deed impossible to imagine, that such a series of signs as those 
 employed in the instruction of the deaf and dumb, and in 
 alphabetical writing, should have been framed by men not 
 previously within the reach of oral communication. A system 
 of arbitrary visible marks analogous to our verbal sounds, as 
 significant of things and ideas, must have been subsequent 
 to the use of language — strictly so called. — It could not have 
 been reasoned out by degrees, for the whole race of mankind 
 must have perished from want before any progress had been 
 made. — It could not have been constructed at a moment, for 
 since nature has given us but very little assistance towards 
 its formation in any instinctive impulses, reason only could 
 have been employed; and it might easily be demonstrated 
 that we owe nearly the whole, certainly all the superior, facul- 
 ties of our intellect entirely to language. — So that the use of 
 words must have preceded that cultivation of the understand- 
 ing, which was necessary to create an ocular language. We 
 are too apt to undervalue the importance of words as an in- 
 strument of thought. — And to consider our voice, with its 
 organs of articulation, rather as a curious appendage to our 
 frame, than as a sixth sense — as the lever, in fact, which raises 
 us above the brutes. — We owe, in reality, not language to rea- 
 on, but reason to language. At least every kind and degree of 
 reason which is not equally shared with the rest of the animal 
 creation. 
 
 (B). This principle which has been generally neglected in 
 ascertaining the significations of words is very important, both 
 
 k3 
 
134 APPENDIX. 
 
 as a clue to dead languages^, and as giving scope for the exer- 
 cise of many valuable intellectual faculties. — A great distinc- 
 tion is of course obvious between this principle and the one 
 which follows it. Analogy would lead us to call the top of a 
 mountain its head^, or the reason of man the light by which he 
 walks. But the application of many very dissimilar ideas to 
 the same word^ arises not from any perception of similarity or 
 proportion between them, but from their being generally 
 united in the same thing or person. A very common illustra- 
 tion may be found in the Greek word ^evoq, which expressed 
 both a friend and an enemy, a host and a guest, a stranger, 
 and sometimes what was beautiful and uncommon. Not that 
 any analogy existed between these ideas, but because in a 
 country where no inns existed, and relations were formed be- 
 tween families residing in different countries, which obviated 
 this inconvenience by mutually entertaining their travelling 
 friends, the same person stood frequently in all these various 
 relations. — And as the domestic economy of the Greeks was 
 on ordinary occasions of an inferior description, and the arri- 
 val of a foreign friend was the signal for greater display, 
 levo? is also employed to signify a superior kind of thing; as 
 distinguished from those in ordinary use. — So also the same 
 word xdpiq is employed to express the agreeable qualities 
 which produce affection, the affection itself, the action by 
 which it is demonstrated, the present made, and the feeling 
 of gratitude which it excites. So also in Latin the word 
 Jldes means the confidence felt, the cause of that confidence, 
 or the honour and integrity of the person in whom it is re- 
 posed — the promise which produces confidence, the adherence 
 to it, or fidelity, and the protection which is promised. So 
 the word 'Ev^a,i[AQvia, is employed by Aristotle, and with great 
 confusion in the result, to denote the feeling in the mind, as 
 well as the circumstances which produce the feeling — and his 
 /A€(70T7j^ is a similar instance — for since one mean equally di- 
 vides two parts, the term is employed to signify equality in 
 general. 
 
 (C). Perhaps in the science of mind, as in many other de- 
 partments of philosophical inquiry, no source more fertile of 
 error could be mentioned than the creation of a technical voca- 
 bulary, before the subject matter was fully comprehended. — 
 
APPENDIX. 135 
 
 The distinction, frequently made in metaphysical works, 
 between sensations, ideas, perceptions, notions, and other 
 words of a similar kind, having once been assumed to be 
 real, from the existence of a distinct nomenclature, has been 
 generally acted upon by writers, and involved in great obscu- 
 rity the inquiries with, which it has been connected. It 
 might seem that all our states of mind are separately per- 
 fectly simple, and incapable of being decompounded — as a 
 single sound, the sensation which we term sweetness, the 
 perception of touch or colour, the feeling of heat or cold. 
 When many perceptions of colours are united with the per- 
 ception of certain muscular actions necessary to carry the 
 eye along lines, we obtain from their succession what may be 
 called ideas. The word idea being properly limited to our 
 perception of figures and magnitudes. — But these ideas instead 
 of being simple are formed by the composition of many conse- 
 cutive perceptions, which appear continuous from the same 
 principle by which a stick on fire whirled rapidly round will 
 to the eye describe an uninterrupted circle. Again, some 
 states of mind produced by action on the organs of sense are 
 pleasing, some painful — and considered in this point of view 
 they may be termed sensations. — But perhaps it is better 
 where so little has been accurately defined to use the words 
 indiscriminately. If it were possible to conceive that our 
 perception of figured bodies was one single state of mind, we 
 might call it an idea. — But it is more philosophical, and 
 better calculated to simplify the rudiments of the science to 
 consider it as a compound operation. — And one thing we may 
 be assured of, that the question can never be solved, at least 
 by our own consciousness — since to obtain a notion of time, 
 and consequently of consecutiveness, we must perceive a cer- 
 tain number of ideas intervening between the two extreme 
 points. Those therefore which follow immediately on each 
 other can never be ascertained to be consecutive — and may 
 appear simultaneous, though not so in reality. The preva- 
 lence of this opinion may perhaps be accounted for by the 
 rapidity with which the eye instinctively glances over objects 
 to collect perceptions, end the wonderful elasticity and rest- 
 lessness of that organ, which, though without our consciousness, 
 is perpetually in motion, even when most it appears to be sta- 
 
136 APPENDIX. 
 
 tionary. And though I imagine that I am simultaneously- 
 listening to the sound of a bell and looking on the fire, it is 
 probable that even these perceptions are consecutive. At 
 least we have many instances where the action of one organ 
 will completely preclude the operation of another. And per- 
 haps these instances may differ from the ordinary state of our 
 perceptions, solely in the duration of the influence^ arising 
 from its character either of pleasure or pain. 
 
 (E). Two or three instances of the metaphysical accuracy 
 of the Greek language may be worth mentioning. 
 
 It is but a short time since philosophy allowed that what 
 we term ideas are merely the mind in particular states. And 
 the expression of forming notions and ideas^ was used as if 
 they possessed an external existence independent of the mind. 
 The Greek vovq, however, signifies the mind, and nothing but 
 the mind, and jivoxtkco is gignere mentem, to form, not an idea, 
 but a mind ; as we say in English, to make up our mind to a 
 thing. 
 
 It was not acknowledged till lately, and perhaps even 
 now the fact may be doubted by many, that all our rea- 
 soning is carried on by words ; that demonstration is conver- 
 sant with nothing but words, and that without the use of 
 language we should possess no reasoning faculty above the 
 brutes. — This very important truth is, however, shut up in 
 the Greek word X070?, which is at once both language and 
 reasoning. 
 
 Again, the whole of Smith's Theory of Moral Sentiments, 
 deriving our feelings with respect to the moral conduct of 
 others from our sympathy with them, is comprised in the de- 
 flective significations of the word o-yyyyol/xvj, to think with a 
 person, to feel with him, to indulge his errors, and think 
 charitably of his faults. 
 
 Again, the use of the verb BaviAoC.a, with three cases, 
 is very precise, and philosophically accurate. With the 
 accusative it seems to signify to admire, to gaze upon a 
 person, eyeing him all over, and feeling pleasure at the con- 
 templation. With the dative, it means to court, to flatter, 
 to fawn upon a man, so as to affect him with certain feelings 
 towards ourselves. But with the genitive it means surprise 
 and astonishment. This genitive evidently signifies the part 
 
APPENDIX. J 37 
 
 of a whole^ the part not being expressed because not de- 
 finitely known. And the most superficial view of the mind 
 in its perception of surprise, will show that it is felt only 
 when there is an incongruity in an object ; when the con- 
 junction is uncommon. That in fact we never do feel sur- 
 prise at any one whole thing, but at some part of a thing. 
 
 Many other similar instances might be adduced; and it may 
 be laid down as a general principle that metaphysical accu- 
 racy is to be found rather in ancient and rude languages, than 
 in others with more pretensions to philosophic correctness. 
 
 THE END, 
 
 TALBOYS AND BROWNE, PRINTERS, OXFORD, 
 
4