(Hmmll Winivmit^ Jihatg THE GIFT OF ja^ .hSAO.m : OKDEE OF STUDY. 17 Heading, as an initiation to tlie knowledge of a language, has a decided advantage over hear- ing* We have a greater command over what we read than over what we hear. In reading we can pause at will, and direct attention to the passages which require investigation ; we can compare what strikes us at the moment, with what precedes, and thus the whole is better con- nected in the mind and more thoroughly under- stood. In Ustening, on the contrary, the slowness of our conception, or the volubility of the speaker, does not always permit us to follow him. TTe have not time to dwell on the words or phrases which call for explanation, and much less can we * The word h^uing commonly means the act of the ear which perceiTes the sonnds ; bat we attach to it the idea of compre- hending the spoken hmgaage. The dirision of our subject and the absence of a special term, justify the second application giren here to this wdrd, by analogy with those which designate the three other arts. Speaking, its counterpart, signifies the act both of uttering articulate sounds and expressing ideas. Reading means both to pronounce the written words and conceive the ideas conyeyed by them. Wriiing appUes equally to the manual art of penmanship and to the expression of thought. The mental operation expressed by the second acceptation of these words is the meaning attached to them throughout this essay. 18 THE STUDY OF LANGUAGES. trace back our steps, in order to compare the dif- ferent parts of the subject and judge of the whole. The rapidity with which most people utter their words in ordinary conversation, re- quires them to be frequently repeated in order to be retained, whereas, by dwelling as long as we please on written words, we always have it in our power to make sure of them as we proceed in reading. The progress made in learning a lan- guage must be at once more certain and more rapid by reading than by hearing. In the mother tongue a child naturally ac- quires the pronunciation subsequently to the meaning of words, and remains ignorant of the written signs, which he is afterward taught by a special process, based on his knowledge of the articulate words ; he passes from the ideas to the sounds, and from the sounds to the letters. The spoken language, the direct sign of his ideas, gives him the key to the written language. In the same manner, but in an inverse order, he who learns a language from books becomes ac- quainted, in the first instance, with the written expression, which is for him, as for the deaf and dumb, the direct sign of thought, and will after- ward, by ear-practice, learn the pronunciation cor- SUBDIVISIOK AITD OEDEE OF STUDY. 19 responding to the written words with which he is familiar. The more perfect his comprehension of the written language, the more rapid will be his progress in the spoken language. When words have been long observed in books, and heard from the teacher's lips, asso- ciated with the ideas they represent, no difficulty will be experienced in reproducing their orthog- raphy and pronunciation, the preliminary ac- quisitions for writing and speaking. The first two arts, considered also as ultimate objects, again claim priority over these, as being, through life, far more usefal. This order is the more rational, as it is much easier to hear or read a language than to speak or write it. The first two arts require of the learner only a slight acquaintance with the words and phraseology ; and, in many cases, their mean- ing is apprehended from the context. In speaking and writing, on the contrary, neither the most acute sagacity nor the greatest in-, ventive power will avail ; not only must we pre- viously know the words expressive of the ideas to be conveyed, but we must also be intimately ac- , quainted with their various shades of meaning, their orthography and pronunciation, their in- 20 THE STUDY OF LANGUAGES. flections, and syntactic or idiomatic arrange- ments. Among the eager crowds who press around a great orator, or among the thousands who read a good author, how few there are, who could speak like the one or write like the other ! In fact, a very limited education will suffice for the com- plete mastery of the arts of hearing and reading ; hut to write and speak well are the fruit of long study, and the exclusive 'privilege of superior intelligence and extensive information. The longest life would not suffice to attain perfec- tion in the last two arts, even in the national idiom. ' In point of usefulness, writing a foreign lan- / guage comes last, and reason suggests that what '^ least needed should be last learned ; we must, therefore, as in the vernacular, place speaking be- fore writing, which is only its representative — the thing signified before the sign. On the other hand, conversation is not at- tended with the same inconvenience, nor is there the same necessity for precision as in epistolary correspondence, or any other kind of composition. Errors in speaking may always be corrected at the time they are committed. A learner is able SUBDIVISION Airo OEDEE OF STUDT. 21 to converse mucli sooner than to write a letter ; because oral expression, aided by tlie language of action, requires fevrer words than written com- position, which, is deprived of this auxiliary. The interchange of ideas in conversation presents more inducements than the elaborate task of epis- tolary composition, and affords greater facilities for acquiring the phraseology of the teacher by imitation and analogy. Let it be well understood that this order, on which we 'insist in the progressive course of studies, does not mean that the learner must be completely master of each of these arts before proceeding to the next ; but that he ought, at the / outset, to direct his attention exclusively to the first object, then divide it successively between that and the other three, as his progress in each makes it an aid to the acquisition of the others. These different objects having become familiar, may be studied together, without any risk of con- fusion. Such is the nature of the human mind : it em- braces a diversity of known objects without con- , founding them ; but two new ones create perplex- : ity, when considered simultaneously. Neither will be known, if not studied separately. " Di- 22 THE STUDY OF LANGUAGES. vide and conquer," a political maxim of Macliia- velli, applies equally to instruction. Tlie order to be followed in tlie study of a foreign language is, therefore, as follows : 1. The art of reading. 2. The art of hearing; 3. The art of speaking. 4. The art of writing. In the first two arts the words recall the ideas ; in the last two the ideas suggest the words. The direct association of the ideas with their signs is what constitutes the real practical knowl- edge of a language, and what, in course of time, forms habits which make it the instrument of thought. This end is attained by a system of judicious repetitions. Repetition is indeed the soul of a good method ; it engenders habit, and habit fixes the acquirements in the memory. Perfectible and communicative as we are, Divine Providence has endowed us with two great instincts which, by contiaually impelling us toward our destiny, create that habit and insure 1 success in the acquisition of language. These are curiosity and imitation. Curiosity, that noble privilege of humanity, that insatiable desire to SUBDIVISION AOT) OEDEE OF STUDY. 23 know, is ever on the lookout for new sensations, for new ideas, and thus enriches the mind : it is the source of progress in the arts of reading and hearing. Imitation, the basis 5f education, iden- tifies us with our fellow-men, and prompts us to adopt their language with their notions, in order to communicate, our thoughts to them : it is the source of progress in the arts of sjpeaJcing and writing. In his mother tongue, a child, whether he listens or speaks, practices the association of ideas with their signs unconsciously, by the mere im- pulse of nature. But, when a second language is studied through the first, it is by comparing one with the other— in other words, by translation — that the learner passes from the known to" the un- known, and attaches ideas to the foreign expres- sion ; some efibrt and a firm determination are then required to adopt the direct association of ideas with words. The first process is the nat/ural or joractical method ; the second, the a/rtifioial or comparative method. The latter cannot of itself insure the perfect knowledge of a language ; but, as it calls for reflection and judgment, it becomes a useful auxiliary for the development of the intellect. 2i THE STUDY OF LAISTGaAGES. The practical knowledge of a foreign living language requires the successive application of these two methods ; for it must have been inter- preted for some time by the native idiom, before its words can be associated directly with the ideas. This double process is the praetical-com- fafabme method, which is chiefly the subject of this essay. All extraneous exercises not tending directly to that knowledge, are only accessories more or less useless, or rather impediments placed in the way of youth by blind routine. The period of learning will be shortened, if the method be sparing of those preparatory exer- cises, which sacrifice the end for the means, and which not only render the student's labor un- profitable, should he discontinue the pursuit, but also divest study of interest, by keeping out of sight the object at which he aims. The special exercises to which recourse is had for acquiring an art, should always be identified with the art it- self, and be its practical application. The yoimg child, left to himself,- rejects theo- ries, and at once avails, himself of all the new ac- quisitions he makes in his own language ; nothing prevents learners from following this example in STTBDIVISION AND OEDEE OF STUDY. 25 anotlier. If, in the task imposed on them, they see something which is really useful to be gained, and constantly in prospect, they will be stimu- lated by the thought that their efforts must have a practical result. Being able to apply the knowledge acquired, as they advance, success be- comes for them a powerful incentive, and a con- tinual source of enjoyment. Thus, a good | method makes the path of duty one of pleasure. / The great secret in education consists in ex.- eiting and directing the will : that system is tlie best which elicits- the greatest amount of volun- tary exertion from the learner. By calling forth all the resources of the student, and making him conscious of his progress, a rational method leads him to incessant spontaneous efforts ; it does not dispense with labor, it directs and seconds it ; it does not impose learning on the memory, it indi- cates the means of acquiring it, of making dis- coveries, and thus renders study accessible to those who are unable to procure masters. The prevailing notion that we must be taught every thing is a great evil. The most extensive education, given by the most skilful masters, often produces but inferior characters ; that alone which we give to ourselves elevates us above 26 THE STDDT OF LAKGUAGES. mediocrity. The eminence attained by great men is always the result of self-imposed labor. On the other hand, exclusive dependence on a professor might proye fallacious in the present state of educational science, . when teaching is purely empirical ; those who select it for a pro- fession do not usually prepare for it by pedagogi- cal studies; and, in their inexperience, they do not always succeed in imparting to their pupils the knowledge which they . po'ssess. To know what to teacJi and how to teach are two very dif- ferent things. This is more especially the case with the teaching of living' languages, a career yjnostly entered upon accidentally by those who, from unfavorable circumstances, have not the choice of a better one. He who feels the want of learning a language, and who, having a definite object in view, is de- termined to attain it, will always rely on his own efforts, rather than on the ability or knowledge of a professor. With a view to direct the unaided efforts of students, we have marked out the task which devolves on them. The professor, on his part, must encourage their spontaneousness, and teach them only what they cannot learn by them- selves. « SUBDIVISION' AND OEDEE OF STUDY. 27 The better to , effect this object, the exercises we recommend are . chiefly those which, by pre- senting good models to the learners, guard them against errors, that only serve to render indispen- sable the aid of an instructor, and are a bar against, good habits: these are formed by the practice of what is right, not by the correction of what is wrong. With regard to children as yet incapable of! self-direction, they must be assisted through the whole course of their studies. But the younger they are the slower will be their progress. It is a strange mistake to think, as many people do, that the great facility with which children acquire their own language is a proof of their aptitude for 1 learning languages in general, and that this studV suits them best. It is not, as commonly believed, because memory predominates in a child, that he masters his language so- easily. This acquirement does not consist in learning words : his attention is engaged with complete propositions, not with in- dividual words ; and yet he firmly retains the latter, in consequence of their frequent recurrence and their association with the ideas on which his mind is bent. The admirable spirit of inquiry 28 THE STUDY OF LAWGUAGB3. Vv'hicli Nature has given to tlie child, is soon cheeked, if we present to him words instead of the ideas he wants. His progress is the consequence of his physi- cal and natui'al condition, which makes his native tongue an object of incessant attention ; he does not receive an idea nor experience any sensation, pleasure, or pain, that is not accompanied by an expression, which is thus engraved on his memory by association. , The practical method, by which we learn our / native idiom, requires only the instinctive exer- cise of curiosity, which calls forth the action of the perceptive and imitative faculties. The com- parative method, which leads to the knowledge of a language through the medium of another, re- quires, on the contrary, the cooperation of in- tellectual powers far greater than are possessed by young children. It is only when the student can command at- tention arid concentrate it on the objects of study, when he can call to his aid reflection and judg- ment, when the maturity of his reason enables him to comprehend serious books; it is then and only then that he can study by himself, and learn a second language through his own. The better SUBDIVISION AED OBDEE OF STtTDT. 29 he knows the latter, the more easily will he learn the former. Under the conviction of this truth, and with a view to laying a proper foundation for the com- parative study of foreign languages, we have dwelt at some length on the course of instruction by which an extensive practical knowledge of the native idiom may be gained during the first period of youth.* From his earliest years, it is true, the child shows a wonderful aptitude for learning lan- guages; but it is exclusively by practice. The constant need, which, from his helpless condition, the child has of those who surround him, and his anxiety to know their thoughts, and enter into communication with them, make it necessary for him to seize on the great bond of union that con- nects him with his fellow-creatures. Thus has the Supreme Being endowed us in early infancy with the inclinations and faculties which gratify this first yearning after social life. If an infant be spoken to in a foreign as fre- quently as in his native tongue, he will become equally familiar with both. He might, in this * See book ir. of " Language as a Keans of Mental Cul- ture," etc. 30 THE STUDY OF LAJJGUAGfES. way, solely guided by nature, learn from the cradle two or three languages without confound- ing them, if brought into daily contact with per- sons who spoke them in his presence, as is fre- quently the case in the higher classes of society, in which children learn the use of several lan- guages. They have governesses and servants of different countries, who always address them each in his own language. Every period of life has its special obligations and occupations which prepare for the next. We must not anticipate the course of nature, and re- quire of one period what belongs to another. Be- fore the age of twelve or thirteen, a child cannot learn a language from books by the aid of his own ; the weakness of his understanding, his want of motives for study, and his reluctance for sedentary occupations, thwart the efforts of the master, who then employs more time in ascertain- ing whether his pupils have clearly understood him, and have learned their lessons, than he de- votes to real teaching. This observation applies more particularly to classical studies ; they are commenced too soon and commenced the wrong way. It cannot even be said, in favor of the early study of a foreign idiom, that it makes a SUBDIVISION AND OEDEE OF STITDY. 31 deeper impression on the mind ; out of a hundred persons who have studied a language by the com- parative method before their twelfth year, ninety- nine have but a faint recollection of it a few years after they have left school. The incomplete knowledge which a young child possesses either of things or of his own lan- guage is, as well as the immaturity of his intel- lect, an impediment to his comprehending foreign authors. He must indeed find it difficult to ren- der the noble thoughts and admirable style of great writers, when, as yet, he has conceived only the simplest ideas, and has at his command only the most familiar expressions. "With the help of a dictionary, he may trans- late every word of a foreign author ; but, in many cases, he will still remain ignorant of his mean-]' ing ; because a dictionary in two languages gives I the corresponding terms without defining them, | or explaining their sigiiification. , The child does no more than render one unknown word by another equally unknown — a baneful practice, which accustoms him to take sound for sense, and disposes him, for the rest of his life, to indulge in empty talk and false reasoning. But, should even a child succeed, with his 32 THE STUDY OF LAKGtJAGES. master's assistance, in rendering the original text with tolerable accuracy, this is not the true end to be attained. It is degrading great writers, ancient or modern, to subject them to a transla- tion which gives the letter, not the spirit, of the original, and to make their noble pages mere parsing-lessons. They are entitled to a more honorable rank. The scope of their works, the wisdom of their views, and the beauty of their diction, ought to be not only appreciated, but imitated in the national language, a task far be- yond the powers of childhood. The important lessons to be learned, and the intellectual enjoyments to be derived from ancient literature, are lost to the mature man, owing to the childish conceptions which he associated with the classics at school, and to the unpleasant recol- lection of all the misery attendant on the study. " The flowers of classic genius with which the teacher's solitary, fancy is most gratified," says Sir "Walter Scott, " have been rendered degraded in his imagination, with tears, with errors, and with punishments ; so that the Eclogues of Yirgil and the Odes of Horace are each inseparably al- lied in association with the sullen figure and mo- notonous recitation of some blubbering school- StIBDIVISION AISD OKDEE OF BTITDT. 33 boy." Such are the pernicious consequences of the premature study of the classics, that Lord Byron, whose mind was so well fitted to enjoy the heauties of Horace, had he read his writings at the proper time, complains in poetical and bit- ter strains of the unconquerable dislike with which the scholastic system had inspired him for that poet. Lamartine makes an observation of the same tendency, in his " Pilgrimage to the Holy Land." "Each wave," he says, "brings me nearer to Greece. I touch its soil ; its appearance affects me deeply, much less, however, than it would have done if all these recollections were not ac- companied by the consciousness that instruction was forced on me to satiety and disgust before I could comprehend it. Greece is to me like a book of which the beauties are tarnished, because I was compelled to read it before I could under- stand it I prefer a tree, a spring under a rock, an oleander on the banks of a river, or the fallen arch of a bridge, covered with the foliage of some climbing evergreen, to the monuments of one of these classic kingdoms, which recall to my mind nothing but the ennui they gave me in my boyhood," 34 TUE STUDY OF LANGUAGES. In order to give a more distinct idea of the principles propounded in the preceding pages, we will recapitulate them in the following maxims, which may be considered as the axioms of our method : 1. Nature is our best guide in the study of languages. 2. To think in a language is the primary con- dition for knowing it. 3. The study of the signs implies the previous possession of the ideas. 4. The association of ideas with their siens results from practice. 5. To practise a language is to receive and ex- press ideas through it. 6. Curiosity and imitation are the source of progress in the study of languages ; hence, (1.) Example is better than precept ; (2.) Practice should precede theory. 7. We must, following the dictates of nature, proceed from the whole to its parts. 8. The means should be consistent with the end. 9. We should never lose sight of the end pro- posed. SUBDIVISION AND OEDEE OF STUDY. 35 10. What we wisli to remember must be con- verted into a babit. 11. One thing only should be done at a time, and every thing in its time. 13. Eeading . leads to hearing, hearing to speaking, and speaking to writing. CHAPTEE II. THE AET OF EEADING. "Lire, lire.et toujours lire dans la langue gtrangero, c'est le moyen par oxcellenco." AjASSONSE GiBAlTSAailE. To read is to conform to one of the laws of our nature-t-the instinct of curiosity; it is to follow on the page the ideas which the writer has consigned to it ; it is to appropriate them, as well as the forms under which he presents them. Books, the depositories of the intellectual treasures which generations bequeath to succeed- ing generations, are the most efficient instruments of instruction in all the branches of literatm'e and science. The variety of information which a proper course of reading brings under the con- sideration of a student, and the opportunities it affords him of surmounting thf, intricacies pre- sented by the different acceptations of words, by technical expressions, and idiomatic forms, will secm'e the means of enjoying the commerce of the THE AET OF EEADIJSTG. 37 well infomied and taking part in their conver- sation. The nations with which foreign languages enable us to exchange thoughts, having diverse origins, living under different climates, brought up in habits and subject to laws peculiar to themselves, must also have ideas and opinions differing from ours. Their wi'iters must see in a different light many questions which have also been treated by our national authors. In history, in politics, in belles-lettres, in the arts, and in other departments of knowledge, their notions often widely differ from ours ; the perusal of their works will therefore enlarge the circle of our ideas and bring us nearer to the truth. In short, the habitual reading of good works in different languages has a most beneficial influ- ence on our understanding ; it stores the memory with knowledge, and leads, by the force of sym- pathy and imitation, to the highest conceptions, and to the practice of all that is great and good. By continual contact with superior minds, we not only come to feel their emotions, to think their thoughts, and to speak their language, but our own sentiments are refined, our thoughts elevated, and our power of expression extended. 38 THE STUDT OF LAITGUAGES. A good book is the best companion of our leisure hours ; we can at any time have recourse to it, and select one from which, as we feel in- clined, we may derive either amusement or in- struction. It is otherwise with men ; we cannot command their services for either purpose, when we are inclined to converse, and it rarely hap- pens that our thirst after knowledge can be satisfied by those we meet in society. "We have not here to treat of reading con- sidered as the art of attaching to the written Words the articulate sounds, which they repre- sent, and which, in the mother tongue, recall to the mind their correlative ideas-; this is not the case in a language of which we do not know the pronunciaticm, nor can the written words lead to it; because the sign recalls the thing signified only to one wh&cJcnows that thing, the art of reading the native, and that of reading a foreign language, cannot be assimilated. The art of reading, which, in the case of a foreign language, is the most useful, both as an end and as a means, may be acquired without the assistance of a master. If, in infancy, we learn, alone, to understand the spoken language, and afterward to speak it, we ought to be able^ THE AET OF READING. 39 in the maturity of reason, to learn, unaided, how- to understand the written language and write it. To teach one's self how to read a foreign Ian- | guage is almost a necessity; for it often happens/ that a foreigner who teaches his own language is not sufficiently versed in that of his pupils to ex- plain the text of authors, to indicate the expres- sions equivalent to those of the original, or to correct the mistakes they make in translating. It is by no means a rare occurrence, that a person, unable to earn a living in his own coun- try, goes abroad with the intention of teaching his native language, without previously preparing himself by learning that of the country where he wishes to establish himself as a teacher, and with- out even thoroughly knowing his own. Gold- smith, the author of the " Yicar of Wakefield," relates how he went to Holland with a view to teaching English there, but found, on his arrival, that a knowledge of the Dutch language, of which he was utterly ignorant, was indispensable for 'the realization of his project. He then thought that the wisest course he could take was to return to England by the very same ship which had brought him thence. An unknown text can be explained only by a 40 THE STDDT OF LANGUAGES. known text, wliicli is its C(|uivalent, and wliich cannot be discovered by one ignorant of tlie laii- gnnge, for we can translate only what ■^^•e under- stand. This equivalent, or translation, a professor naust know perfectly, in order to present it vint, voce to his pupils when very young. A printed translation will suffice for students who are old enough to dispense with a master. Grammar affords no assistance in reading ; it does not explain the meaning of phrases or words, which is the only difficulty encountered in learn- ing to read a foreign language. The translation which interprets the unknown text, not the gram- matical condition of the words, must be the first, the only object for the beginner's consideration. Grammar may teach a person who speaks or writes incorrectly, how to speak or A\'rite cor- rectly, but it certainly is not the ai-t of re<((ling and understanding a language. No one has ever been insane enough to attach to it this definition. This twofold acquisition is, as we have clearly shown, the first thing to be mastered by students : grammar, therelbre, is obviously iiscloss at the entrance on the study. Syntax, more especially, cannot bo an auxil- iary, inasmuch as, contrary to the order of THE AET OF BEADING. 4-1 nature, it puts precept before example, theory before practice, and makes ns pass from -n-ords to tlieir combinations. In tlie progressive develop- ment of the intellect, the perception of an object al'n-avs precedes the consideration of its parts ; we learn to understand our own language by passing from the phraseology to the words. The latter have no value but that which is assigned to them by the phrase. The function they perform in speech determines the class to which they belong, as well as their signification. Eesorting to the phrase for an explanation of the words is to pro- ceed from the idea to the sign. If we then wish to follow the prescriptions of nature in acquiring the art of reading a foreign ' language, we must not prepare for it by learning | either its grammai" or its words. The latter canj be known only in books, by means of the context which fixes their signification. It is not by a previous study of words that we come to under- stand what is said to us in our own tongue ; on the contrary, it is by hearing and reading that we form a vocabulary for ourselves. " Is not," says De Gerando, " the nomenclature of a language, taught as a preparatory exercise, whatever care may be taken, most uninteresting, and hence 42 THE STUDY OF LANGUAGES. most prejudicial to the first stage of the study, when it is so important to make this first stage easy and attractive ? " He will, however, make an exception in favor of a very limited class of words. The elements of speech which form what we / call the fiurst class of words, are the substantvoes, adjectives, and viei'bs. They are significant by themselves, and constitute the essential parts of a proposition. The other elements of speech, forming the second class, the articles, pronouns, prepositions, adverbs, and conjunctions, merely serve to con- nect, modify, and complete the sense of the other three species of words.* * We do not include inia-jedions in our classification ; because, different from the other elements of speech, they are not conven- tional terms, and have not any fixed character ; they vary in their pronunciation and application with the temperament of individuals, and the different emotions which give them birth. As instinctive cries of nature, they are universal, and belong to the language of action, not to any particular idiom. It is then an error to class these inarticulate sounds among the parts of speech, especially as they never enter into the construction of sentences, and are not subject to any syntactical law. Laughter, shrieks, and all other involuntary vocal convulsions, might as justly be called parts of speech. These remarks apply to interjections properly so called, not to THE AET OF BEADING. 43 To these words we will add .the expletives, whicli have been denied a place in grammatical nomenclature, although acting an important part in discourse. They serve to point out the gram- matical condition or function of the words before which they are placed, such as to, the sign of the present of the infinitive mood, and shall, the sign of the future. II (it) in French, is an expletive, when it marks the impersonality of the verb. It is to the second class of words that the, above-mentioned exception refers. An -acquaint- ance with them might be useful, if gained at the outset, or studied simultaneously with the prac- tice in reading ; for, although apparently of secondary importance, they are the binding links of discourse, and materially modify the sense of sentences. Committing to memory words of the first <3lass would not assist a beginner ; . for as their difier- ent acceptations depend on a phraseology as yet unknown, the vagueness of their import in a vo- cabulary would occasion difficulties. But, besides these different acceptations, their other essential exclamatory expressions, such as, away I bravo ! heaven ! hark ! help ! murder, etc., which, although ranked among interjections, are elliptical propositions, formed from different classes of words. 3 4:4 THE STUDY OF LANGTJAGES. elements, pronunciation, orthograpliy, inflections, syntactical concord, and place in the sentence, constitute a series of considerations wliich the at- tention could not embrace at the entrance upon the study. The absence of connection between these words in vocabularies must also make it difficult to retain them. Even if they were remembered, they would be of little avail in translating ; for different subjects, different styles having terms peculiar to each, and rarely to be met with in works to which they do not properly belong, the student might not meet in his first volume with many of those which he had been at so much ipains to learn. It is from the connected discourse in which they are incorporated, that their precise import can be ascertained. In imitation, there- \fore, of what occurs in acquiring the vernacular, these words must, in the foreign tongue, be learned, by reading and hearing, and not be made the means of learning these branches. The words of the second class, on the contrary, enter into all compositions, whatever be the sub- ject or the style ; most of them having only one signification, acquaintance with it will facilitate the understanding of the text. The uniformity THE AET OF BEADING. ' 45 of tTieir orthography, and their very limited num- ber (not 400), will render their acquisition easy ; while their frequent recurrence in every thing the student reads, will fix them firmly in his memory. Let any one open a French duodecimo volume, and he wiU find that every line contains four^ five, or more words of this class. In number they form a hundredth part of the other words'; but, in composition, they occur twice as offceji ; so that each word of the second class is used on an average two hundred times oftener than one of the first class. A. previous acquaintance with these words will prove useful, especially in reading French ; be- cause, as their orthography differs completely from that of the corresponding terms in English, their meaning cannot be divined like that of substan- tives, adjectives, and verbs, three-fourths of which bear great resemblance to each other in the two languages. As the means should always be consistent with the end, it will sufl6.ce to study their form, so as to be able to recognize them at sight. The same may be done with the inflections of the verbs, and those of substantives in languages which admit of cases. This is not learning grammar : the words 46 'the studt of laitgttages. of the second class, the declensions and conjuga- tions, considered apart from any technological distinction or syntactical functions, but only as regards the ideas they express, appertain to lexi- cology, not to grammar. However, in the ab- sence of a special collection of them, they may be found in their respective places in all the gram- mars which give the classification of the parts of speech. At this early period of the study, and for the object now in view, it would be worse than useless to commit these words to memory, as the student would thereby infallibly acquire a bad pronunciation. Out of fifty syllables formed of the same letters in French and in English, there is scarcely one which represents exactly the same sound in both languages ; so that, in pronouncing French words, which he has never heard, the learn- er is likely to be wrong forty-nine times out of fifty. If, however, owing to the habit of always pro- nouncing in his own language the words which he sees, he cannot help doing the same with those of another language, he must, when ignorant of its pronunciation, be content to give mentally to the letters of the foreign words the sound which they have in his own language : this will be, as it wei'e, THE AET OF EEADING. 47 a mere spelling, not an attempt at pronunciation. The Tocal organs, remaining, in the moan while, quiescent, will not contract any habit which must be got rid of afterward. Direct reading, that by which the written ex- , pression, as in the native idiom, directly conveys the thought, is the end to be attained. Indirect reading, that by which the idea is apprehended / through the medium of the mother-tongue, that is, translation, is only an introduction to direct reading. At an advanced stage of the study, translation becomes an obstacle to the understand- ing of the language, for it is not always possible. So would spelling, which is an auxiliary in learn- ing to read one's own language, prove an obstacle to reading if persevered in. The process by which a learner is enabled to follow directly the train of an author's ideas can- not be made too easy. This spontaneous associa- tion of the ideas with the words that represent them constitutes the real practice of the foreign language, and the groundwork on which ad- vancement in all its other departments is based. Translation, on the contrary, being the practice of the national idiom, becomes thereby a most efficient means of improvement in it. 48 THE STUDY OF LAIfGUAGES. But, as direct reading can be arriTsd at only through the medium of translation, the student must, as a preliminary step toward it, attend seriously to the latter. No parsing, no gram- matical comment on the language : all he requires is to advance rapidly in the comprehension of the text in hand, that he may become acquainted with a large number of words and phrases. Prac- tice is now the object ; we will subsequently sug- gest modes of mental culture. The first books to be used should treat of • familiar subjects, and be written in an easy style, in order to ayoid encountering at the same time the difficulty of the subject and that of the lan- guage. Attention is then directed to the form, not to the matter ; it is absorbed in the work of translation. All .serious instruction, apart from the language itself, would be iU-timed. "These books are, as it were, only practical or reading vocabularies, but vocabularies addressed to the understanding as well as to the memory, and the words of which have a definite meaning. They will familiarize the student with the terms and phraseology of ordinary conversation, at the same time that they will lay the foundation for studies of a higher character. The elements of THE ART OF BEADING. 49 discourse, by the daily practice of reading, like all daily occurrences, remain in the memory -with- out effort, as deposits from the stream of experi- ence. Correctness of language is nearly all that i^ required in the text of these initiatory books. Any other merit would be lost, at least out of place, at a period when it cannot possibly be dis- tinguished, stiU less appreciated, especially as the meaning is reached only by translation. The initiatory texts to be translated should be rather below than above the age of the learners, who should never be required to read works which would be above their comprehension if written in the national idiom. In the study of languages, as in that of the fine arts, master- pieces are not fit for beginners ; novices always i work on materials of an inferior kind. The most ( eminent writers and orators have, in childhood, passed through the ordeal of trivial language and commonplace ideas. "We insist on this point, because the prevalent notion that none but works written in the most elegant or classical style ought to be put into the hands of beginners, creates the necessity of resort- ing to various preparatory exercises, and is in op- 50 THE STTTDT OF LAJSTGUAGES. position to the principle of gradation dictated by nature : it is one of the chief causes both of the discouragement experienced by learners at their entrance upon the study, and of the unreasonable duration of linguistic instruction. Modern literatures present inexhaustible food to curiosity, and great facilities for the strict ap- plication of the principle of gradation of difficul- ties ; they abound in books which, being intended for young people, may serve as an .introduction to reading, and as models for learning to speak- and write. In this respect, living languages have a great advantage over the dead : in the latter, the number of works which have been handed down to us is too limited to permit in all cases the difficulty of the task to be adapted to the learner's capacity. Their elevated character is beyond the reach of immature minds ; it is there- fore the understanding of the learner which, by proper delay, must be raised to the standard of the classics. What is above the capacity of a boy of ten or twelve may have some chance of being understood by a lad of fifteen or sixteen. The learner must read a considerable quantity of prose before he enters on poetry. It is by gradual steps that, in the native tongue, we are THE AET OF EEADING. ' 51 enabled to commune witli superior minds. It is absurd to make poetical compositions of a higli order a means of study ; tbey are its end, its re- ward. He who uses tbem as a means, will not feel a wish to read tbem when he has learned the language. What shall we say of the common practice of putting into the hands of young people who have read only a few volumes, the works of Dante or of Milton, poets whom their own coun- trymen can scarcely understand ? * Nor are voluminous works fit for beginners ; they lose all their interest, on account of the slow- ness with which they are read. The kind which appears to combine the most favorable conditions are first the books which- treat of subjects familiar to the student, then fables, anecdotes, tales, nar- ratives, and historical sketches ; these are subjects of general interest, and become more interesting stiU if they relate to the nation whose language * " Clii oramai in Italia, chi & che reramente legga, e intend a, e guBti, e vivamente senta Dante e Petrarca ? TIno in mille a dlr molto." — ^Alfieki. The style of Milton's " Paradise Lost " had become so anti- quated, so obscure, about a hundred years ago, that a bookseller, named Osborne, thought proper to publish a prose version of it for the benefit of " ordinary readers." 52 THE STDDX OF LANGUAGES. is studied. A style intelligible to children and purely narrative, as being the easiest, is the most suitable for beginners, even those of mature in- tellect. Leatners, whatever be their age, will apply the mors willingly to this task, as it grati- fies instinctive inquisitiveness, and is the least painful of all that are imposed by the study of a language. By facilitating their first steps in read- ing we command their attention, give them a taste for reading, and secure their success. The reading of the foreign text may be com- /menced at the outset, without any preparatory studies or exercises, by means of a literal transla- tion. With the interpretation of that text before his eyes, the student, having first perused an Eng- lish phrase, will then utter it 'with his eyes directed to its foreign equivalent ; that is, he will translate the latter in mentally attaching, as far as it is practicable, the known to the unknown words. For greater facility in passing from one text to the other, these should be placed opposite to each other in the first books which he uses. In all countries possessing a literature, the best works have been translated and are daily trans- lated from one language into another. There can be no difficulty in procuring these auxiliaries. THE ABT OF EEADtSTG. 53 but, out of the great number of translations, students must select those which, as they render faithfully the foreign text, most closely follow its construction. Their great merit for a beginner is to be literal, and yet written in a clear and cor- rect style. The previous knowledge of the words of the second class and of the inflections of those of the first, which we recommend above, will be the more useful according as the auxiliary transla- tion is less literal. The mode of interpretation which we-recom-'^ mend is peculiar in so far as it permits a foreign language to be studied through an English trans- lation, or an English original text which has been translated into that language. The same book may, therefore, serve for students of the two coun- tries. All the works, for instance, published in France and in England or America with the French and the English opposite each other, are equally useful to the English or Americans who learn French and to the French who learn Eng- lish. These interpretations, by removing uncer- tainty as regards the true meaning of the foreign text, far surpass in efficiency the usual mode of translating with the help of a dictionary, which continually leads to errors that call for assistance. 54 THE STUDY OF LANGITAGES. They dispense with the necessity of either a master or a dictionary. The enormous time con- sumed by the latter, and the perplexity arising from its various interpretations, discourage be- ginners, and delay their progress, when they have to look out for nearly all the words of their au- thor, s Words, moreover, which are thus translated one by one, present but a vague meaning, and fre- quently none, to a child as yet little versed in his own language. Their signification depends on the very text the sense of which he is seeking. To find out the unknown through the unknown, such is the circle in which he is placed by the dictionary. It is partly owing to this illogical, repulsive, and unnatural process that must be attributed, for the great majority of young persons, the signal failure of linguistic studies. "With the aid of a dictionary they hardly translate, and translate badly, twenty-five or thirty lines a day, about a volume in the course of a year, whereas twenty- five or thirty volumes at least should be read to secure the complete acquisition of the art of read- ing. Some people imagine that the use of the die- THE AET OF EEADING. S5 tionary impresses the words on the memory, for- getting that this mode of coining at their mean- ing is not the fruit of reflection, and does not con- stitute a discovery, any more than being told it or taking it from a translation : it is a mere rehance on the testimony of others, with the additional un- certainty and confusion arising from various inter- pretations for the same word. Its inefficiency as a mnemonic auxiliary is proved by experience. The languages learned by drudging at it, are mostly forgotten with amazing rapidity. How could it be otherwise, when the use of thumb and fingers is substituted for the exercise of the Intel- ■ lect ? The native words gained without the dic- tionary are retained with extreme tenacity. In any case, that pretended auxiliary is incompatible with the principle of our nature which leads us to pass from the whole to its parts, from the phrase to the words. The dictionary, however, can be made use of when, at an advanced period of the study, the learner meets with few unknowi words, and is able from the context to choose the right meaning from among those given in its col- lumns. The readiness vsdth which a learner through a translation in juxtaposition, seizes the thought of 56 THE STUDY OF LANGUAGES. the author, and the logical sequence of the sub- ject, gives an interest to reading, which it cannot have with the dictionary, as the latter, by direct- ing his attention to each word individually, breaks off the connection between the ideas. If the annexed translation enable the learner to get through the text more easily than he can with the dictionary, he will translate more in a given time, so that the same expressions will present themselves the more frequently, according as they are more useful, and, as in the mother-tongue, will be remembered in proportion to their useftdness. Progress is always in an inverse ratio tolhe time devoted to translating the first volumes. One hundred pages, for instance, read at the rate of ten pages a day, will advance a learner more than the same number would, if read at the rate of one page a day. The foreign phraseology to a beginner is a true chaos, in which the eye perceives nothing distinctly : by degrees, the frequent reappearance of the same elements in their appropriate places exhibits their essential characteristics and makes the object of attention gradually clearer. After a while, the light begins to dawn, and they all present themselves in intelligible and harmonious order. THE AET OF BEADING. 57 The division of the labor which falls to the share of the master and pupils respectively, is favorable to public teaching, as it allows the lat- ter to advance in reading in proportion to their desire of learning and to the time they can spare for study. The quantity which they may read daily should be regulated, not by what an in- structor has leisure to hear in the class, but by the time which they can devote to it in their pri- vate studies, and by the facility with which they perform the exercise. All the members of a class, according to their different degrees of proficiency, may read differ- ent works, especially as the professor does not make these a subject of examination for each pupil separately. It is, as will be seen, when he comes to initiate them in the art of hearing that he can judge of theif diligence in his absence. In this way, diligent learners are not kept back in their studies by the indolence or inca- pacity of some of their school-fellows, as fre- quently happens in the present state of public teaching, in which the professor imposes the same book and the same task on all, without the least regard to the intellectual inequalities that may exist among them. This mode of reducing the 58 THE STUDY OF LANGTTAGES. intellects and capabilities of all to the same stand- ard, condemns some to a deplorable inactivity, and others to a task above their strength. The mode of proceeding at the commencement should be nearly as follows: To devote exclu- sively to the translation of the first volumes all the time one has for study in the absence of the teacher, to go several times over the same pas- sages for some weeks, to peruse every day the iesson of the day before, and gradually throw off dependence on the translation opposite. As the work becomes easier, more will be translated in a given time, and the learner will soon be able to dispense with auxiliary aids. He should, however, guard against excess in this respect, against prematm-e attempts at perfec- tion. By dwelling too long on the first pages the task would be made tedious and disagreeable, at the very time when his curiosity needs to be stimulated by variety and novelty. Besides, by such dilatory minuteness, words and phrases of rare occurrence would be apt to occupy time and attention to the exclusion of those which are more immediately required; whereas, by steady prog- ress through the book, he wiU more frequently meet with those which are the most useful, and THE AET 01" BEADING. 59 his acquisition of tliem will be consistent with the demands of colloquial intercourse. Besides, we can know the full import of words only hy meet- ing with them in various circumstances. We pro- ceed thus in the native tongue and in all the arts. Long concentration of the mind on one subject, like the division of labor in manufactures, creates habits which impair its power. Although the ordinary practice of teaching has not hitherto favored the publication of works in two languages, there are, nevertheless, a suffi- cient number in English and in Erench to initiate learners of either nation in the art of reading the language of the other. Dialogues and collections of phrases in the two languages might even be used for this purpose in the absence of the books recommended. "When, with the aid of two or three of these works, the student has become familiar with a large portion of the foreign phraseology, if not yet able to dispense altogether with assistance, he may have recourse to books in which the more difficult expressions are explained in notes at tl^p foot of the page. Several works of this kind may be ob- tained for the French and the English languages. Interlinear translations — among the advocates 60 THE STUDY OF LANGUAGES. of which may he mentioned the great scholars of Port-Eoyal, Dumarsais, Beauzee, Kadonvilliers, Condillac, D'Alembert, and Locke — certainly render an important service, as auxiliaries in reading; we therefore do not absolutely reject them, especially if they are accompanied by a free translation. But, as they make the words subservient to the meaning of the phrase, in op- position to the principle we have laid down, we prefer interpretation on the page opposite to the foreign text, which explains its phraseology. In- terlinear translations, in addition to the confusion for the eye, arising from the jumbling together of the two languages, are a constant cause of per- plexity from the diff'erence of their construction, and the more so, when the words, as it often hap- pens, are always translated the same way, what- ever be their meaning in the text. As the student advances in the comprehension of the foreign text, he will naturally notice the words which the phrases have in common, and will the more readily apprehend then- precise import, in proportion as they often recur and in difi'erent circumstances. JSTot only the words but their prefixes and aifixes, as well as their in- flections, will be explained one by the other. THE AET OF EEADING. 61 When, at a later period, having laid aside the initiatory hooks, he meets with new words, their points of resemblance, to those he already knows, their roots or their terminations, the place they occupy or the circumstance that introduces them, will be so many data to lead him by induc- tion to an inference as to their meaning. ISTot only would this investigation be favorable to mental discipline, but the information thus gain- ed would be more indelibly impressed on the mind, precisely because it had been discovered by mental efforts. If, on thus appealing to his judgment, he does not succeed in discovering their import, he can then have recourse to a dictionary ; and, in this case, to arrive the sooner at direct reading, we should give the preference to a dictionary, exclu- sively in the language he is learning, which, as it presents the definitions of the words, is less likely to lead him astray than a dictionary in the two languages, which often presents by translation only an approximative sense, "We gain the knowledge of native words by instinctive analysis. The first sentence we hear conveys to the mind an indistinct notion of the meaning of a word, the second makes this notion 62 THE STUDY OF LAITGUAGES. / somewhat clearer, a third and a fourth render conjecture still more definite, until at length, a last induction removes all doubts as regards the idea to be attached to it. In this manner we come, by almost imperceptible steps, to know the precise meaning of a considerable number of ab- stract terms, which no definitions could ever make us understand. Nearly all the words that we know in our own language, have been divined in this man- ner. This is a mental operation far superior to reliance on a dictionary, which is, after all, a mere mechanical operation. What is discovered by mental effort, is more thoroughly known and better retained than what is learned from a book or from a teacher. Three months ought to suflice, without any very great labor, t-o read five or six small volumes, and even to read them twice over. Then, as greater facilities in reading -make it a more at- tractive occupation, the student will read more, and will advance toward perfection with rapid strides. These results, due to the diligence of self-taught learners, cannot, however, be expected from chil- dren whose age requires the aid of a teacher. THE AET OF EBADUSTCr. 63 Their advancement in reading depends on the time he devotes to them, and will necessarily be slov7, especially when, as in public schools, only two or three hours a week are allotted to the teaching of the living languages. It is incumbent on learners to finish a work once begun ; let the instructor recommend none but such as are worthy of being read entirely. Much of the interest and profit is lost, when books are but partially read. The second part K' a work generally indemnifies us for the trouble we have had in reading the first. As we advance in a volume we become acquainted with the author's peculiarities of style, and our minds are gradually identified with his. Perseverance in completing the work, necessarily brings a repeti- tion of the same words and phraseology, and thus engraves them in the memory. The comparative facility also with which the latter part is read is a source of pleasure, and a manifest indication of improvement ; it is unreasonable to deprive stu- dents, as is often done, of this gratification, of this stimulus to further exertion, by making them read only portions of works. More unreasonable still is it to expect that learners can become con- yersant with the literature of a country by the 64: THE STUDY OF LAIifGlTAGES. study of extracts from various authors, however judicious their selection. The merit of a good book, moreover, does not depend exclusively on minute details of style ; it also consists in the end which the author had in view, in the conception of the general plan, and in the harmony of all its parts. " If a book be worth reading once," says Ben- jamin Franklin, " it should be read twice." We will add, if, at an advanced stage, it is not worth reading twice, it ought not to be read at all. A second reading is indispensable both for advan- cing in the art of reading, and for retaining the materials of conversation. It is indeed impossi- ble, on the j&rst reading of a book, to perceive all its force and propriety of expression, or even to attend to the orthography or arrangement of the words, the attention being then engaged by the effort required for understanding the text. On a second or third perusal, familiarity with the matter and the rapid association of expressions and ideas, enable the reader to divide his atten- tion and bestow part of it on the composition of words, their fitness, and their arrangement. Some read much, who yet write very incorrectly ; be- cause they attend exclusively to the subject, and THE AST OF EEADIKG. 65 never bestow a thought on the form of language. If the course we suggest be adopted, orthography in particular, will present no difficulty ; the fre- quent recurrence of the same words in reading will render the eye an instinctive judge in spell- ing, as is the educated ear in pronunciation. The impressions which are made by the first acquaintance with standard works are usually; confused ; it is only on closer acquaintance with them that the mind acquires the power of perceiv- ing the connection of the parts, the character of the whole, the suitableness of the style to the thoughts, and all the beauties of composition, as well as the inaccuracies which have escaped the author. The repeated perusal of a work furnishes the surest means by which its literary merit can ; be tested ; for productions of sterling worth afford, new pleasures, and unfold new beauties at each successive readJiig, whilst those of inferior char- acter scarcely bear a second perusal ; they exhibit more imperfections, according as they are more frequently or attentively read. If a person, when reading alone, meets with passages he does not clearly understand, and can- not, at the moment, obtain the explanations he needs, let him mark them with pencil in the mar- 66 THE STDDT OF LANGUAGES. gin, he will generally find that all the difficulties disappear on a second reading of the volume: such is the fruit of practice. • Eepetition is the grand principle on which de- pends the efficacy of the processes required for gaining a practical knowledge of a language. To impart to the intellectual powers a certain free- dom of action, repetition is as necessary as exer- cise to the limbs. Six months of continuous ap- plication will lead to greater proficiency than twelve months of lessons with frequent interrup- tions. Habits of language can be created only by keeping the same words and phrases in rapid succession before the mind : the same number of impressions which, when closely following each other, produce a habit, would fail to do so, if sepa- rated by long intervals. The reading of the great writers should be de- ferred until it can be effected without the medium of a translation. It is only by direct reading that the mind, free from considerations apart from the subject, can enter fully into the author's meaning. JSTeither scientific nor philosophical works can be studied with the same advantage by translating as by direct reading. A search after the native ex- pressions corresponding to those of the original THE AET OF BEADING. 67 breaks in constantly upon the connection of the subject; and the mind, thus diverted, cannot easily follow a train of close reasoning. The qualities of style which constitute the chief merit of works of imagination are entirely lost by an extempore translation in which the form is necessarily neglected for the substance, the mind being exclusively engaged in rendering the identical ideas of the foreign text ; poetry, es- pecially, cannot be read through translation. All that constitutes its beauty, its charm, disappears in passing into another language. iN'o two languages correspond word for word, phrase for phrase : they all differ in their genius. Each has numerous expressions without equiva- lents in another, and consequently ideas which cannot be exactly represented in .the latter. Hence, the scarcity of faithful translations. There is some truth in the Italian proverb, "Traduttore traditore." "ISText to a good tragedy," says Yoltaire, "nothing is more diffi- cult to write than a good translation." " Of all books," says also Lamartine, " the most difficult to be written is, in my opinion, a translation." The best of them, those which are made by eminent writers, and meditated in the silence of 4 68 THE STUDY OF LAKQUAGES. the closet, are, for the most part, but imperfect copies of the master-pieces which they are in- tended to represent. What must be the extem- pore translation of students, who as yet do not know the foreign language, and frequently are only smatterers in their own ? When the reading of the foreign language is the only object pro- posed, it is a waste of time and useless trouble to polish the phraseology corresponding to that of the original text. If the latter is understood, the end is gained. What signifies the manner of ex- pressing the same ideas in a second language? We will even go further ; it is not the translation which leads to the exact sense of the foreign text, but the clear understanding of that text, which secures the means of translating it properly. With regard to irregularities of construction in idiomatic phrases, the student will do well to be satisfied with the interpretation of his text, as given by a good translation. Much must at first be taken tor granted, consistently with that vital principle, "practice before theory." Inquiring into the reasons of the peculiarities of a foreign idiom only impedes progress through the book without making it more intelligible. Let the reader reflect that, in his own language, he can THE AET OF EEADING. 69 seldom solve difficulties of tMs sort : he could not account for innumerable anomalies and idiomatic forms; although, in common practice, he would properly apply every expression, and would never hesitate about their signification. Few English persons, for example, even among the well-edu- cated, know or care to know the reasons of the following deviations from grammar or from the proper meaning of words : two dozen ; a few salmon / a irace of snipe, of partridge, etc. ; many a day / now-a-days / metliinks / would 1 were there ■ now j were I to put up with it ^ I had rather stay • you had letter he off ; and a thou- sand other equally odd expressions in daily use. The case is just the same with the idioms of other languages. The chief obstacle whicli translation opposes to the complete possession of the foreign lan- guage, is owing to the fact that, being placed be- tween the foreign expression and the thought, it prevents their direct association, and consequently the instantaneous suggestion of one by the other. To be conscious of a sensation and retain it, the attention must be directed simultaneously, with the action of the organ, on the object of sensation. In translation the foreign text, it is true, is before 70 THE STUDY OF LANGUAGES. the eye, but the learner's whole attention is given to the forms of the langnage into which he trans- lates; the translator thinks in the language in which he has embodied the thought, and so the foreign expression leaves no trace in his mind. This process, the one usually insisted on, engen- ders a habit which excludes the possibility of thinking in that language and retaining its phrase- ology. Here we find an additional reason for dis- pensing with the services of a master in this first stage of the study, since he cannot assist his pupils ■ in understanding a foreign author, save through translation. The oral translation of several volumes having familiarized the student with a considerable por- tion of the vocabulary and idiomatic phraseology of the foreign language, he will drop the practice of translating, in order to take in the ideas directly .from the author. A few efforts - in this direction will rid him altogether of this inefficient process, and enable him to follow the ideas in the text itself — ^the first step in the art of think- ing in the language. It is especially on the sec- ond perusal of a book, or of passages of a book, that he must make his debut in direct reading, on account of the facilities offered by his knowledge THE AUT OF BEADING. 71 of tlie text. After a few days' practice he will find that it is far easier to follow the thought of an author, directly than to translate. Until now pronunciation has been set aside, as aJGEbrding no aid toward the signification of words, the reverse of what happens in. the native idiom, which is acquired by the immediate association of the sense with the sound, and in which the written word is intelligible, only inasmuch as it recalls to the reader a known articulate word. Before the foreign words had been frequently heard, bad habits of pronunciation would be con- tracted, if a student gave utterance to them as he reads. The time given to the practice, if he aimed at a correct enunciation, would be lost for his advancement in the comprehension of books. Signification, not pronunciation, is the first step to be made in the study of a language. Three things are to be considered in a word, the written form, the articulate sound, - and th-e idea ; the written form is the sign of the articulate sound, as the articulate sound is the sign of the idea. Now, the sign cannot be an ob- ject of consideration, unless the thing signified be present to the mind; we must, therefore, be thoroughly impressed with the ideas before at- 72 THE STUDT OF LANGUAGES. tending to the articulate sounds which represent them, and must be acquainted -with the sounds, before we turn our attention to the letters. He who is denied the benefit of a teacher's as- sistance, contents himself with pronouncing in his own language the expressions corresponding to those of the foreign text, and thereby guards against a false pronunciation. But, when famil- iarized with the sense of the foreign words, the sounds of which he does not know, he wishes to practise direct reading, the surest means of suc- cess will be, as he peruses the text, to utter the words mentally in the manner recommended at page 46. The mind thus engaged will be diverted from translation. Defective as is this miental pronunciation, the vocal organs not being engaged in producing it, no bad habit is contracted, and the self-imposed error will rapidly vanish under I the reiterated impressions of the true pronun- \ elation, when the circumstances are favorable for acquiring it. Should the learner have the benefit of a teacher, he will master the pronunciation in time to apply it to direct reading, if, by the exercises which we shall hereafter describe, his progress in that art has kept pace with his advance THE ART OF HEADING. Y3 in the understanding of the written language, When he is master of the pronunciation he will always attach it to the text, and will thus make it the immediate expression of his thought. The pronunciation of the words in a low voice, while the mind attends to the subject, will form the or- gans to habits that will aid considerably in using correct language afterward. Direct reading must be diligently practised, in the absence of the professor, throughout the course. It is impossible to become familiar with all the words and the phraseology of a language, in fine, to understand it like one's own, except by the constant and studious reading of good writers in prose and verse on a variety of subjects. Such is the true, the only way of becoming acquainted with the literature of a country and the genius of its language. To know a language, observes the learned orientalist. Sir "William Jones, " we must read an infinite number of works written in it." Of the 50,000 words, or thereabout, comprised in the vocabulary of a modem language, there are at least 20,000 which should be recognizable by the eye and ear, as coming within an extensive practice in reading or hearing, while 10 or 12,000 ought to be known so as to be readily suggested 14 THE STUDY OF LANGUAGES. for the expression of thought in all the circnm- stances of life. It is obvious- that this twofold object can be attained only by a eoui'se of diversi- fied reading, embracing all the subjects on which we may have occasion to converse in society. We do not speak here of the exchange of i-deas which may take place in the occasional inter- course with foreigners whom chance brings in our way : a few words will suffice for this object. We have considered it our duty to show the way to perfection : every one is at liberty to stop when his object is attained. In addition to the difficulty of attaining the ready use of so great a number of words, we must not forget that most of them are taken in different acceptations, a fact which imposes a fur- ther task on the memory, and renders extensive reading still more necessary. ISTothing but long practice can bring under the reader's notice all the idiomatic combinations, all the circumstances which illustrate these changes of meafling. The more extensive and varied the student's readinsr, the more copious will be his vocabulary, and the better will he understand the words in their vari- ous acceptations. It is in order to gain familiarity with the THE AET OF BEADING. 15 idioms especially — a very numerous class of -ex- . pressions not accessible through rules — that we must have recourse to the extensive reading of popular -works in the absence of social intercourse with foreigners. The memory is enriched by the repetition at- tending long practice, and the elements of dis- course, thus associated with ideas in the mind, constitute the first condition for speaking a lan- guage, as well as for entering into the spirit of its writers. The power of this association is such, that he who should practise it with perseverance, especially in his native tongue, by uttering the language of good writers as he reads them, could not fail to possess a rich stock of ideas with a great facility in extempore speaking. There can be little doubt that many good public speakers have been indebted for their success to this mode of self-training. Considered as the means of fixing the elements of speech in the memory, reading is indisputably efficacious, even in the native idiom ; though, in this case, the habit of familiar and sometimes trivial language is an obstacle to the impressions received in reading well-written works. In fact, great precision of expression is only to be attained 76 THE STUDY OF LANGUAGES. by the perusal and the imitation of good writers ; for, in books alone, do we find the well-chosen terms and the forms which characterize a noble and graceful style. The influence of books is more decidedly felt in a second language; be- cause the expressions of the foreign author are not exposed to the antagonism of others already known; they strike the mind with all the force of first impressions. Even those who learn a lan- guage solely with a view to speaking it, ought to read a great deal. In order to be conversant with all the resources of reason and language, we must necessarily seek them in many books. The practice of reading, by exercising curi- osity, multiplies its energy, and people desire to read, in proportion as they have formed the habit of reading. In accordance with this desire they can always procure books, and thus retain the art of reading to the end of a long life; whereas the double talent of speaking and writing a foreign language is very soon lost for want of practice. The art of reading foreign languages, if gen- erally cultivated throughout the civilized world, would greatly facilitate international relations: every one writing in his own language would THE AKT OF EEADIKG. 17 then be understood abroad. Diplomatists, scientific men, and mercbants, especially, would derive in- calculable advantages from tbis attainment ; tbey would cease to be dependent on interpreters and clerks, wbo often write so inaccurately as greatly to perplex tbeir correspondents. International communication has, until tbis time, been mupb impeded by the extreme difficulty of writing a foreign -language. Very seldom could a pereon be sure of conveying his meaning in it, with as much clearness and precision as in his own ; and if be had correspondents in various countries, it is more probable that they could read his language than that he could write their different idioms so as to be perfectly understood.* * A student might sometimes be impeded in his progress tow- ard perfection in the art of reading, by the difficulty of procuring all the books indispensable to an extensive course. Every obstacle of this kind would be removed, if the professor would collect a small library composed exclusively of good works in the language which he teaches, with and without the interpretation of the text, and furnish them to his pupils for a small quarterly payment ; he would thus have the means of directing them in the choice of books best suited for their use. CHAPTEE III. THE AET OF HEAEING. " It is a mistake to Relieve that study and practice are not as neces- sary for liearing as for speaking." Plutaeoh. The advantages arising from social intercourse consist more in receiving than in communicating ideas. Hearing is truly the better half of conver- sation : it is, in every respect, incomparably more useful than speaking. Like reading, . it satisfies instinctive inquisitiveness, one of the principles of human perfectibility. If we perfectly understand what is said, a few words, a monosyllable, even the slightest motion of assent or dissent, will suflBce to keep up conver- sation or transact business. If, on the contrary, we do not clearly comprehend the person who addresses us, all the command of language which, we may possess will be unavailable for social in- tercourse : hence hearing may be useful inde- pendently of speaking ; whereas speaking is use- THE AET OF HEAEING. 1Q less without hearing. It is always more profitable and less daDgerous to listen than to speak. Especially when we are in a foreign cQuntry for the first time, every thing is matter of curi- osity : we have a great deal of local information to gain from its inhabitants, a thousand oppor- tunities of hearing their language, and compara- tively very little to say. If a person understands the spoken language of the country which he visits, he is enabled, from the moment of his ar- rival, not only to enjoy the society of the inhabit- ants, but also to improve in speaking ; for words and entire phrases are easily retained, when the ear distinctly catches and the mind clearly appre- hends them. In the absence of this faculty, he remains isolated in the midst of his fellow-men, and even keeps away from their society, in the dread of exhibiting his ignorance. So that he derives neither profit nor pleasure from a resi- dence abroad ; and he might be years in the country without being able to speak its language. " He that travelleth in a foreign country," says Lord Bacon, " before he hath some entrance into the language, goeth to school not to travel." The comprehension of the spoken language, although the most important element of our social 80 THE STUDY OF LANGUAGES. relations, and, at the same time, the most natural and most certain means of learning to speak, has, nevertheless, been so far neglected in the teaching of foreign languages, that there is not even a term by which to express it ; and, for the purpose of classiiication, we have been under the neces- sity of attaching this new acceptation to the words hearvng and audition. Many persons do not even suspect this power /to be an art; for they have no recollection of having ever learned it in their own language, so easy does nature make its acquisition at the en- trance into life. The capability to speak implies, in the vernacular language, that of hearing, which, by the force of imitation, gave it birth. This is not the case with languages learned from books. The difficulty generally experienced in under- standing oral expression is not inherent in the na- ture of the art : a child sixteen or eighteen months old is already a proficient in it, though he would be utterly incapable of improvement in any other department of language. The difficulty may fairly be attributed to the methods of teaching hitherto followed ; they leave the organ of hear- ing in complete inaction ; or only exercise it on THE AJKT OF HEARING. 81 detached words and phrases. The ear untaught by the teacher's voice, cannot, in the usual- rapid- ity of speech, recognize the foreign words, how- ever familiar they may be to the eye. The learn- ers are exercised in hearing only in conversation, when they begin to speak. This is putting the cart before the horse, since we cannot joia in con- versation, unless we understand what is said. This acquisition is further impeded by the learner's not having previously met in his scanty reading with the phraseology of those with whom he comes in contact ; while, from his want of prac- tice in hearing the language, he is unable to as- sociate the ideas with the sounds, and is then obliged to translate what he hears, an operation for which the rapidity of speech does not allow time. The prescriptions of the preceding chapter tend to remove these obstacles. The student who has followed them, and made some advance in the art of reading, wiH find himself in a favorable position for advancing ia the second branch. If the art of reading can be acquired without a teacher, it is otherwise with hearing and pro- nouncing, in which no advance can be made without his aid, and especially without his pos- 82 THE BTUDT OF LASTGUAGES. sessing a correct pronunciation and accent. When a child pronounces Ms own language incorrectly, it is the fault of those around him; when the learner of a foreign language pronounces it badly, it is the fault of his teacher. The simple and natural method by which the ear is formed to the articulate _ sounds of the mother tongue, and by which the Tocal organs learn to reproduce them, is equally applicable to a foreign language. If strictly and perseveringly follqwed, it would enable us to seize the meaning of the words, and acquire the foreign pronun- ciation as easily as our own. That this end is so seldom attaiaed, only proves the fallacy of the methods pursued. "We disdain to follow the easy path which nature has marked out, and are pun- ished by fatigue and disappoiutment. The natm-e of the object proposed sufficiently shows the mode of procedure. If, in the absence of the master, the eye is exercised on orthography, in his presence the ear must be exercised on pro- nunciation ; these two organs afford mutual aid to each other. Eeading and hearing must be carried on simultaneously. A passage of a foreign author being selected, at every lesson, by the professor, from what has THE AET OF HEAEIITG. 83 previously been studied by the learners, he dis- tinctly reads it aloud in sliort phrases, wMcb they alternately translate without seeing the text, and the length of which is commensurate with their degree of advancement. Then the whole passage is read a second time without interruption, but slowly enough to permit the learner to translate mentally as he proceeds. Being assisted in these two exercises in hear- ing by the recollection of the subject, the recent impression of which is still fresh in their minds, the learners will easily recognize the words which are familiar to the eye. They may perhaps, at first, only make guesses ; but these, by their re- currence, will soon become real knowledge, as is the case with every thing that is learned by experience. The association of the spoken with the written words will be rapidly effected, because the ele- mentary sounds are few in number, and, in con- sequence, recur frequently; moreover, it will take place naturally, and without any of the diffi- culties attending the illogical process by which the sounds are usually inferred from the letters in teaching the foreign pronunciation. The professor will avoid reading isolated 84 THE STUDY OF LANGUAGES. words ; for their signification, and especially tliat of homonyms, that is, of words differing in mean- ing but resembling in sound, can be understood, on being heard, only from the sentences in which they are incorporated. If, in reading to a class, the passage selected be the last prepared by the least advanced learner, while he translates it chiefly from recollection, his more advanced fellow-students, who may not have read it for some time previously, will also do the same, owing to their greater familiarity with the language. Thus, all the members of the class, whatever be their different degrees of advancement in reading, wiU derive equal benefit from this exercise. In this manner the professor always has it in his power to ascertain the diligence of his pupUs in that part of the language which devolves on themselves. As he reads to them for. translation the passage which they had to read in his ab- sence, he thus examines them in one branch while exercising them in another ; for the pronunciation will not suggest the sense to his pupils, unless it recalls to their minds a text previously studied. And when he wishes to ascertain the progress they have made in orthography by due attention THE ART OF HEAEING. 85 to the written form of the foreign text in pre- paring tlieir lessons, he will stop at the words, which present some difficulty, and will question them on their spelling, accompanying this ex- amination with explanations and rules which will benefit the whole class, however numerous it may be. This will preclude the necessity of dictations and other orthographical exercises, from which no advantage is derived in proportion to the time they consume. The reading by fragments will be set aside, and the second, or uninterrupted reading, con- tinued for a longer time at each sitting, when the pronunciation, having grown familiar to the hearers, does not call for great efforts on their part, and thus permits them to attend easily to continuous discourse. The professor will then lead his pupils from the known to the unknown, by reading to them what they have not previously seen ; then, by a gradually increasing rapidity of utterance, he compels them to associate the idea with the sound, to pass from translation to the direct comprehension of what is said. The art of following ordinary conversation presents no difficulty to a person able to under- stand the language on hearing it read; for the Ob THE STUDY OF LANGUAGES^ words used in tlie exchange of ideas are generally- more familiar than those wMcli are found in books, the periods are shorter, and the same ex- pressions more freq^uently repeated, while the de- livery, more natural, is accompanied by tones, looks, and gestures, which greatly assist the hearer. The mind is kept alive by the ever- varying topics, and relieved by the successive in- terruptions of colloquial intercom'se. The person spoken to is also more attentive, because he feels more interested in what is personally addressed to him. This truth is forcibly illustrated by the well-known fact that, in public assemblies, ex- temporaneous speeches are much more favorably received than written discourses. The professor will accustom his pupils to the forms of conversation, by choosing from the pas- sages read to them, the idiomatical expressions in common use, modifying them in the same manner as is done in familiar intercourse. He may also make such additions, retrenchments, substitutions, transpositions, as his fancy suggests, so as to change the ideas without altering the essential character of the idiom. As regards his readings, he will gradually lengthen them and make them more rapid in THE AET OF HEAEING. 87 proportion as his pupils, "becoming more familiar "witli the pronunciation, shall begin to understand him without translating. These exercises will soon enable them to understand the spoken lan- guage directly, if, while seizing the thought it conveys, they repeat it mentally, as it falls from the professor's lips. In the mean time the professor will, as ex- plained in the following chapter, communicate to them his pronunciation and accent, and the more easily in proportion as he has trained their ears by much reading. At an earlier period, exer- cises in pronunciation would have been prema- ture : it has no foundation to rest upon, unless the unport of the words is perfectly tnown, to which it is to be attached. How, indeed', can signs be studied, which do not suggest the things signified ? It is a great mistake to imagine that, in the study of a living language, the pronunciation should be taught first. It does not in any way facilitate the understanding of the written words ; and, besides, a person may perfectly understand what he hears, without being able to pronounce correctly. In infancy we know the meaning of words, long before we can utter them. In learn- 00 THE STUDY OF LANGUAGES. ing a foreign language we onglit also to under- stand the spoken words before attempting to ar- ticulate them. To study simultaneously both the pronuncia- tion and the signification of words at the begin- ning is incompatible with that law of our mental organization, which forbids attention to be di- rected at the same time to several distinct things when new. In the course of the exercises in audition, the learners should forbear looking at what is read to them, that the ideas may be exclusively received through the articulate words, as when listening to a speaker. If a person familiar with the written language had his eyes fixed on the book while the. instructor was reading, that organ, quicker than the ear and not easily controlled, would not always patiently accompany the reader word for word, but woxdd outstrip the ear in ap- prehending the subject. Sometimes also a per- / son less advanced, will be slow in following the teacher, or will stop to consider the words which are not familiar to him ; so that, in either case, the learner would be unmindful of what is read, and the idea would be apprehended through the eye,, not through the ear. THE AET OF HEAEIKG. 89 The learner also, occasionally perceiying let- ters wMcli are not pronounced, wonld be apt to attribute Ms not bearing tbem to inattention or dulness of bearing on bis part, and migbt still be inclined to introduce tbem in bis pronunciation afterward. It is, therefore, better not to give the eye an opportunity of leading the ear astray. Besides, this dependence on the sight for under- standing what is beard, incapacitates the ear for conversation, in which it can have no assistance &om the eye. By the reading of familiar subjects, suited to the age and proficiency of his pupils, the master will render tbem the same service as those who initiate us in childhood in the knowledge of our own language. But his mode of reading should be made a perfect fae-simile of that of speaking, by a natural and expressive delivery. Good read- ing is that which most nearly approaches ex- temporized speech. The interest with which a sMlful instructor may invest this exercise by his manner of reading and the choice of subjects cal- culated to excite curiosity, will powerfully contrib- ute to fix the attention of his hearers. He may ' always secure this point by calling on them, from 90 THE STUDY OF LAJSTGIJAGES. time to time, to repeat or translate ttie last phrase he uttered. The learners must now, while listening to the professor, mentally repeat after him. This direct association of the ideas with the sounds is the second step in the art of thinking in the language. The association of the spoken words with the ideas which they represent, offers no difficulty if the hearers have been accustomed to mental or direct reading ; for it is impossible to follow a speaker and think with his words, unless the sounds awake the ideas. Direct hearing is therefore more neees- \ sary than direct reading ; the latter operation is / optional, the former is indispensable. . The fact that persons who, while able to read a foreign language, are yet unable to speak it, arises in a great measure from their not having contracted a habit of the association of words with ideas, by means of much practice in direct reading and hearing. Perfection in audition, which consists in being able to understand oral discourse, however rapidly spoken, will be easily attained, if, " by assiduous study, one has learned to read the foreign like the national authors, as quickly as the eye runs over the text ; for the organs of speech utter the words THE AET OF HEAEING. 91 more slowly tlian tlie eye peruses them. It must be borne in mind that the hearer has not, like the reader, the option of dwelling on an expression ; he must apprehend the ideas of the speaker as they are delivered : he is completely at his mercy. To reach this perfection, a sine qua non in serious conversation, the pupils must desire the teacher to slacken or hurry his pace, as there is •occasion. They should never hesitate to inter- rupt him, when they do not understand what he reads. The connection of the ideas will not suffer, as after every interruption the professor wiU re- 1 sume his reading at the place where he stopped. The rarity of these interruptions will be a sm'e criterion of the progress of his pupils. It would be otherwise if the instructor should prematurely address them in the foreign language; for their interruptions would be so many obsta- cles to his treating a subject connectedly. Thus, thwarted in his attempts, and frequently losing the thread of his discourse, he would soon be compelled to desist; whereas, with a book, he can always suit the simplicity of the subject and - the slowness of his delivery, to the inexperience of his hearers. Moreover, reading renders im- provement in the second branch independent of 92 THE STXIDY OF LANGUAGES. the oratorical powers of a teaclier, wlio may be Tery deficient in this respect, at the same time that it familiarizes them with a better choice of words, a more correct phraseology, and a greater diversity of expression, than can be done in an extempore discourse. He could thus also store their minds with useful knowledge, and accustom them to all styles, to all modes of delivery, without himself possessing any other talent than that of reading. A. practice no less injudicious than that of prematurely addressing learners at some length in the foreign language, and one frequently re- sorted to, although at variance with the order of nature, is for the teacher to draw his pupils out into a conversation, when they are as yet unable to understand him. lifone but commonplace ideas could be ventured upon, and, owing to the difficulty of the attempt, considerable time would be lost ; hence, very inadequate practice Tjoth in hearing and in speaking. More objectionable still is conversation in a large class ; for only a few of its members could seriously join in it, to the exclusion of the greater number, who, meanwhile, would remain com- pletely idle. By reading, on the contrary, the THE ART OF HEAUrNG. 93 professor can initiate in tlie comprehension of tlie spoken language fifty persons as easily as one. His discoTirse is addressed to all, and profits all. Tliose who have not had much practice will en- deavor to catch the meaning of what they hear ; the others will go further, and their attention will embrace both the pronunciation and the meaning. In any case, this exercise affords in a given time more practice in hearing than conversation. Such is the merit of this process; it places the acquirement most essential for the exchange of ideas within reach of those whose limited re- sources will not allow them to obtain it except in large classes. It is well known that, when a great number of persons attend the same course of instruction, all cannot make the same progress : difference in age, aptitude, and taste, as well as in previous education, and the time they have at their disposal, wiU cause some to advance more rapidly than others. It is therefore important that the method should provide for this inequality among pupils of the same class. As poetry does not admit of being translated extempore, the reading of it by the professor will, as it were, force the pupils to associate the thoughts directly with the words, and, at the 94 THE STUDY OF LASTGUAGES. same time, will be for them tlie best practical initiation into' the prosody, if lie carefully mark in reading the syllabic quantity and the tonic ac- cent. In this, as in every other department of the study of a language, practice should precede theory. It is only when the ear, by long ex- perience, has been made conscious of the exist- ence and nature of tones and quantity in syl- lables, that the mind can investigate melody and rhythm, that explanations can be given as to what constitutes the essence of verse, and the mechanism of the foreign versification. The understanding of the spoken language in Italian, Spanish, and German, presents great facilities, owing to the correspondence between the pronunciation and the orthography. But the most difficult of all languages for a foreigner to understand is perhaps the English, on account of the complete absence of analogy in the alphabet- ical representation of its pronunciation, as weU as of the rapidity with which it is spoken and its in- numerable contractions.* * This is humorously illustrated in the following anecdote : In a late trial before the Queen's Bench, Mr. Hawkins, a barrister, had frequently to advert to that description of vehicle called THE AET or HEAEING. 95 Some people think that the French language is spoken fastei: than the English ; this is a great error. Yoltaire shrewdly observed, that an Eng- lishman gains eyery day two hours on a French- man in conversation. The truth is, that English is spoken considerably quicker than French. This results from a difference of kind in the pro- nunciation of these languages. Pronunciation is composed of two elements, vocal sounds and articulations, represented in writing by vowels and consonants. Yocal sounds admit of duration: quantity is their essence. Vocal articulations, with few exceptions, cannot be prolonged; instantaneity is their essence. When a consonant is placed after a vowel, it generally shortens it. Thus the long syllables, brougham, which he pronounced in two syllables. Lord Camp- bell, the chief justice, suggested that the word was usually con- tracted to broom, and that he had better adopt the latter pro. nunciation, as he would thereby save one syllable and gain so much time. Henceforward Mr. Hawkins called it broom. Shortly after, the pleading turned upon omnibuses ; and Lord Campbell frequently used the word omnibus, to which he gave its due length. " I beg your lordship's pardon," retorted Mr. Hawkins, " but, if you will call it bus, you will save two syllables, and make it more intelligible to the jury." The learned judge assented to the proposed abbreviation. 96 THE STUDT OF LAI^GPAGES. me, we, fie, no, due, though, become short by add- ing consonants to them, met, web,fib,fit,fi>g, not, dun, thought. ISTow, in English, consonants pre- dominate, and usually form the end of syllables ; hence a rapidity of utterance is the unavoidable consequence. In French, on the contrary, consonants act but a secondary part, and are often silent. The spoken words, in reality, end with vowel-sounds, although consonants terminate their written rep- resentatives. In the division of the words, con- sonants seldom terminate syllables; the French word ccMficature, for example, is divided into syl- lables thus, ca-ri-ca-tu-re ; its pronunciation, con- formably to this division, is necessarily longer than that of the English word, commonly pro- nounced, according to this other division, car-ic-a- ture. The same may be said of every other word in the two languages. The vowels, which con- tribute so much to lengthen the wordsj are pro- nounced full in French, as if every syllable were accented. From these facts there necessarily re- sults a slow and steady enunciation. As the opinion of a foreigner, however, in re- gard to the English pronunciation, can have little weight, we beg to quote a few competent au- niE AET OF HBAEING. 97 thorities : " Such, is the vehemence of our accent, that every syllable which follows the accented is not only short, but almost lost in the pronun- ciation." (Lord Monboddo.) ""We incline, in general, to a short pronunciation of our words, and have shortened the quantity of most of those which we borrowed from the Latin." (Hugh Blair.) "Such is the propensity for dispatch that, overlooking the majesty of words composed of many syllables aptly connected, the prevailing taste is to shorten words, so as to make them dis- agreeable to the ear." (Lord Karnes.) " It must be regretted that contraction subjects our tongue to some of the most hissing, snapping, clashing sounds that ever grated the ear of a Vandal." (John Walker.) " Our rational conversation is, for the most part, carried on in a series of most extraordinary and rugged abbreviations, a species of short-hand talking." (Bulwer Lytton.) But, whatever be the irregularities of the pro- nunciation or the rapidity of speech of a people, it is the business of learners to overcome the former, and accommodate themselves to the latter. Prac- tice wiU suffice to enable the ear to distinguish the most delicate shades of sound ; and the pro- cesses recommended above secure this object. 98 THE STUDY OF LAJSTGtTAGES. The professor who exercises his pupil in hearing has this great advantage over the mother who initiates her child in this department of the lan- guage, that the person he addresses, having a more mature understanding, must necessarily ap- prehend the ideas corresponding to the sounds moi-e quickly and clearly than can be done by a young child through the language of action with which his mother accompanies her words. He owes his success to her admirable patience in re- peating the same phrases, always associated with that language. The professor has, in translation and the previous reading of his pupils, no less powerful elements of success ; and, if he exercises the same patience and the same perseverance in his teaching, he wiU obtain the same results in considerably less time. Perseverance is equally indispensable, in order to preserve the acquisition. A language is, in fact, retained not in proportion to the degree of perfection attained, but to the length of time it has been practised, and to the strength of the habits contracted by the ear. The youngest chil- dren who are taken abroad lose their own lan- guage most rapidly, that is to say, according as they have heard the less of it. THE ART OF HEADING. 99 "When the language is perfectly understood, the professor will turn the proficiency of his pupils to aecoumt by always using it in speaking to them, even before they themselves speak it. The con- stant practice of following a train of ideas dh'ectly through the medium of words spoken by a native, will render the audible signs so familiar as soon to secure a habit of this mental operation ; and, this once attained, it will cause these ideas, in virtue of the laws of association and habit, to be easily reproduced when the hearer has afterward need to convey them. The pronunciation, thus directly associated with the thought,- and daily becoming a firmer habit of the ear, the attention may be directed, while following the ideas, to the articulate sounds which represent them, as a preparation for the exercises in speaking, described in the following chapter. The faculty of reproducing the articu- late sounds of a language, is the consequence of the frequent impressions made by them on the : organ of hearing. Those are dumb to whom nature has refused the sense of hearing. The accent, especially, which afifects the whole tenor of discourse and consists in a vocal modula- tion peculiar to the nation, is fostered in the 100 THE STUDY OF LAUGUAGES. hearers by the reiterated impressions received from the foreign teacher's voice, if he reads as the language is itsually spoken. The longer the ear is impressed with the national accent, the more its fibres vibrate in unison with it, and the great- er is the power of the vocal organ to assimilate with it. By the mode of proceeding which we have described above, a person possessing the art of reading a foreign language, would be able in six /weeks, or two months at the most, to understand perfectly those who speak it, if he had at his com- mand a reader, whether a professor or not, who \ should, for half an hom* at a time, at short inter- ' vals, read to Mm the foreign language : and if he visited the country where it is spoken, he would have still greater facilities for understanding its inhabitants and adopting their pronunciation ; a servant, a child even, who knows how to read, could rapidly forward him in this twofold ac- quirement. Success is, in any case, accessible to the most humble fortune; for the first art is learned without a master, and the second requires his services only for a very short time. These two arts are so easily and speedily ac- quired, when, conformably to the laws of nature THE AET OF HEAEIKG. 101 and of reason, their study is freed from all the i fetters of routine, that one might in less than six . months read" and understand French, for instance, ' as well as a native. He could not, however, in as many years, learn to speak it like a Frenchman. It is, therefore, most desirable that, among all civilized nations, the attention of youth should be more particularly directed to the first two arts, which, if universally diffused, would suffice, to the exclusion of the other two, for all the require- ments of the international exchange of thought. Persons of different nations, each speaking or writing his own language, would understand each other, and their conversation would be the more expansive, the more satisfactory in every respect, as, in ordinary circumstances, ideas flow in the vernacular with more freedom and clearness than in a foreign idiom. In this way would the grand desideratum of modern society, international ex- change of ideas, be secm*ed. The reciprocal knowledge of living languages cannot fail to extend our aocial relations and to render international intercourse more frequent and more useful; it would second the work of civilization, by promoting the progress of the arts and sciences, doing away with national prejudices. 102 THE STUDY OF LASTGUAGES. and drawing closer the bonds wHch ought to iznite all the members of the great human family. The literary and scientific celebrities, who popularize useful knowledge and new discoveries in public lectures, would be easily induced to visit neighboring countries, as they would then have every opportunity of collecting around them numerous hearers able to understand them. We do not, indeed, see why lecturers should not be patronized by an enlightened public, when foreign actors frequently perform to crowded auditories, l^Teed we expatiate on the many advantages which, in a moral, intellectual, and social point of view, would accrue to society from this exchange of information and good offices. Never was a common means of intellectual communication more needed than at the present day. Different communities, despite their rulers, tend to fraternize under the influence of similar institutions, similar pursuits, and similar tastes. The spirit of the age impels nations to form political and commercial alliances on all points of the globe, and to blend themselves into one great community. Scientific associations succes- sively attract to the great centres of activity all THE AKT OF HEAEING. 103 the noblest intellects of the civilized world. Iso- lated labor is everywhere giving way to the spirit of association ; and instead of wrapping their dis- coveries in secrecy, men of all countries diffuse them as means of universal advancement. CHAPTEE lY. THE AET OF SPEAKING. "Ici rapplication serait meilleure que les regies, les exemples iu- Btmiraient mieux que les prSceptes." Bukpos". " Here practical application would he better than rules, examples more instructiye than precepts." As, iu the present state of linguistic education, the first two arts, reading and hearing, are not /universally diffused, the third, the art of speaking / the foreign tongue, becomes invaluable for every / person who comes in contact with foreigners un- : able to understand his language. It is especially the case, after a foreign idiom has, lite the native, become the direct instrument of the mind, that it promotes intellectual culture. The exchange of thought is, in fact, an essential element of improvement. In social intercourse, when a diversity of characters are brought to- gether, every one contributes his share of knowl- edge, good sense, and experience, and all are THE AET OF SPEAKENG. 105 gainers to some extent. If reading enriclies tlie mind, conversation polislies and expands it. In conversing with those whose esteem we covet, or whom we would fain convert to onr own way of thinMng, the desire to please, to persuade, keeps all the faculties of the soul in a state of excite- ment, which multiplies the intellectual energies, and often leads to the conception of ideas, which would never have been evolved in the solitude of the study. It would, therefore, he wrong to neglect an art capable of producing such impor- tant results. By reading and hearing, the student familiar- izes himself with the models ; by speaMng and writing, he imitates them. The habit of receiving ideas directly from the words, written or spoken, lays the foundation for rapidly acquiring the faculty of expressing them spontaneously. The art of speaking thus finds infallible elements of success in the practice of the first two arts. But, without waiting until perfection is gained in these, the learners wiU pass on to the exercises of the third. "With this view, they must imitate the pronunciation of their professor, and the phrase- ology of standard authors. The task of the instructor, as regards pronun- 106 THE STUDY OF LANGUAGES. ciation, should consist less in correcting Ms pupils' errors tlian in preventing tliem from com- mitting any. Prevention is better tlian cure. / As a general rule, they ought never to pronounce / a word, unless they have heard it several times. I So long as the organs do not practise any pro- ' nunciation, they remain ready to acquire a good one ; but a bad habit once contracted, is eradi- cated with difSculty. There will be no danger of falling into a de- fective pronunciation, if, following the law of nature, as manifested in infancy, the pupil listens to his master a long time before attempting to imitate him. The exercises of the preceding chapter, having formed the ear to correct habits, the vocal organs will be in a favorable state for giving the true pronunciation. The vocal faculty is governed by the ear ; there are no articulate sounds perceptible to the organ of hearing which the voice cannot reproduce. Pronunciation is most commonly taught by imposing on a beginner the oral reading of the foreign text, that is, making him infer it from ' the orthography — ^a mode of proceeding doubly irrational, as it rejects imitation in an art exclu- sively based on this power, and as it implies that THE ABT OP SPEAKING. 107 the tMng signified, not previously known, can be learned from its sign. It moreover submits tbe pronunciation to the learner's attention without- regard to the meaning of the words, although their pronunciation often varies with their signifi- cation. Tor instance, the pronunciation of the words l)ow, gill, read, desert, gallcmt, conjure, rebel, mvnute, and many others, depends on the sense in which they are taken. It is the same in other languages, though the English, especially, abounds in words which differ in pronunciation, according to their meaning. Reason dictates that the means should be con- sistent with the end ; but reading aloud is pre- cisely the reverse of what takes place in conver- sation. In speaking, we pass from the idea to the word ; the sound suggests the orthography : in reading, on the contrary, we pass from the word to the idea ; the orthography suggests the sound. Eeading aloud can only be a source of errors for a beginner. The correction to which it leads could not create good habits : these result alone from reiterated correct impressions, such as arise from the processes explained in the foregoing chapter. Its proper office is to test the progi-ess made in pronunciation. 108 TIIIO STUDY OV LANOtJAaiOS. Tlie two readiiif^'fl by tlie inaBlcr, as explained in' the procoding oliM.ptor, wore intondoil to ini- tiate Iho loiuTicrs in tlio undcrriliuidinf;; of the Bpokon languiT^'o, luul to accustom the ear to a fi;ood proniinciiitioii. IIo will now make thom iiiitcr on the praci ico of the latter hj reading; the same passiiiiicrt twice more — the first time A'ory slowly, and by short ]ilu"asc9 of tlirco or four words, which tlioy will repeat after him without lookiuf!,' at the text, and in imitation not only of the sounds, prosodial accent, and blondiiia; of the words, but also of the intonation and inflection of voice jjeculiar to the people. Those imitations should, as in a(',(piiriiig the native tongue, be made without reference to the alphaboticnl characters. Whonovor the pupils lail in reprodueiiip; the exact j)ronuiioiatiou or ac- centuation of their model, it is a jti'oof they need to hear it again. The professor will, there Core, at each failure, ntter the woi'ds anew. The voice, that docile slave of the oar, in order to echo the pronunciation faithfully, only needs to have it clearly impressed on tliin orij;iut. In his second reiidiiifj;, which will be uninter- ru])tcd, he will set the eximiplo of a correct de- livery by an accentuation conloi-mable to the THE AET OF SPEAKES-G. 109 habits of speaking of ■well-informed people. His hearers, previously familiarized with the spoken language, haTing now heard the same passage read fonr tim^ in snecession, will experience no difficnlty in associating mentally the pronnncia- tion with the ideas. TVlien, hr means of these imitations, a stu- dent has a complete mastery of all the element- ary sounds of the foreign language, he will direct his attention to the alphabetical signs which rep- resent them, and will find it both easy and inter- esting to attach to the written words, as is done in the maternal idiom, a pronunciation perfectly lamiliar to him. He may then occasionally readi the same passage which the professor has now' read four times, and, in doing so, will direct his attention exclusively to the ideas, in order to de- liver the text naturally and with the inflections of voice required on the subject. The pronuncia- tion is not known until it has become so fixed a habit, tliat. in speaking or reading, it is produced spontaneoiisly and withont diverting the mind from the thought. ^e will here observe that these four readings are, in their nature and effect, precisely identical with what takes place, as already seen, in the first 110 THE STDDT OF LAlTGTTAGES. four periods of our apprenticesHp to the vernac- ular idiom. The child first seizes on the ideas with the aid of the language of action, which ac- companies the short phrases addressed to him ; he is soon able without this assistance to xmderstand contiuuous speech on matters within the scope of his intelligence ; next, he tries to pronounce the _ words and phrases he has most frequently heard ; and, at last, when, by speaking, the pronuncia- tion has grown familiar to him, he is taught to read. Of the four readings given by the professor, the first and third, in detached phrases, being in- ; tended only to initiate the pupils, the first, in comprehending, the third, in pronoimcing the language, need not be continued for any length of time. But consecutive reading wiU be perse- vered in, in order to confirm the learners in good habits of pronunciation : according to their prog- ress, this reading will become more rapid and be continued longer at a time. "When, both from hearing and from reading, accurate impressions have, by repetition, grown familiar to the mind, any deviation from them, which one may afterward meet, would strike him as being incorrect. This consciousness of pro- THE AET OF BPEASIING. Ill priety, arising from the habit of the ear or the eye, as formed in good society or from good boots, is a practical conviction, a true experimental knowledge, and our best guide in the expression of thought. Good speakers and good writers are guided by the ear rather than by rules. It may sometimes be desirable to commence the exercises of pronunciation at an early period of the study. There is nothing to prevent it, al- though it is never advisable to act contrary to the dictates of nature ; but, at whatever period they are entered upon, it is indispensable to pronounce only words which are fully understood, and have been frequently heard. In a language the orthography of which faith- fully represents the pronunciation, the written and the articulate words, being easily inferred one from the other, will not only render reading and hearing mutual auxiliaries, but vsdll enable the student to arrive, by analogy, at the pronun- ciation of the whole language from that of a few words. Eeading aloud, unavailable in most languages, as a means of learning pronunciation, will serve to keep it up at a later period, when once per- fectly acquired. A good habit of pronunciation 112 THE STUDY OF LANGUAGES. /may likewise be kept tip by storing the memory ,' with select passages of prose and poetry. The possibility of repeating them at any time and in any place offers an easy means of obtaining that correct and natural enunciation which is the crowning perfection in the pronunciation of a language. "When reiterated imitation shall have enabled the learners to reproduce faithfully and naturally the pronunciation of their model, that is, when it shall have become habitual to them— the professor wiU lead them to another kind of imitation, that of the foreign ^phraseology — an exercise the easier as the attention is free from any consideration of pronunciation. The arts of speaking and writing are acquired / by the same process as that which leads to excel- lence in other arts. Imitation and practice can alone in all arts produce a good execution. It is as irrational to make grammar the starting-point in leamiag to speak, as it would be to impose on a child the study of perspective or the theory of colors as the preliminary step to learning the art of painting. In both, practice, founded on ex- ample, is the basis of improvement. Practice alone may, by induction, lead to a knowledge of THE AS.1 OF SPEAXnTG. 113 grammar ; tlie latter can never of itself lead to practice. If, as already shown, the knowledge of the second class of words is a useful aiixiliary in learning to understand the foreign text, it must be equally useful when we wish to express our thoughts. Those are, in fact, the words which enable us to vary the phraseology indefinitely. The professor ought, therefore, to have them be- fore him, when exercising his pupils, that he may readily introduce them into the sentences which he gives them in their own language for construc- tion in the foreign one. The learners, on their part, when thoroughly masters of the pronuncia- tion, will find no difficulty in committing them to memory, as they become familiar by frequent re- currence in reading and hearing. A previous knowledge of the verbs is equally useftd in the first steps toward the art of speaking. This word par excellence, constituted, as it is in cultivated languages, with- its moods, tenses, per- sons, numbers, and inflections, is the most in- genious of instituted signs, the vital element of discourse, and the masterpiece of language. It expresses in itself a judgment and a proposition. Its inflections are, in some languages, so numer- 114: THE STUDY OF LANGUAGES. ous tiiat a very long time would be required to know tliem all by practice only. It is, therefore, useful to make them an object of study, cMefly witb a view to learning the different relations which they express. Let it be remembered that an acquaiatance with words or the import of their inflections does not imply the study of grammar, which more properly consists in definitions, rules, and disquisitions on language. The verb is truly the essence of the proposi- tion : without it we can neither affirm, deny, nor question ; it will necessarily form part of every sentence, which the master gives his pupils as models, to exercise them in the phraseological variations. We can speak from the moment we know one verb perfectly, and can unhesitatingly apply it in every possible manner without refer- ence to the mother-tongue. If it be considered that a verb in all its moods, tenses, and persons, and in its various forms, active, passive, and reflective, affirmative, inter- rogative, and negative, presents above a tJwusand distinct propositions, it wiU clearly appear that, by successively joining to it the other elements of speech, an iaexhaustible diversity of expressions" may be produced. By changing the words of the THE AET OF SPEAEmG. 115 second class, we obtain different modifications of the same idea ; and, by varying tbose of the first, we express different ideas under similar circumstances. The thought undergoes at pleasure an endless metamorphosis, and the vocal organs acquire cor- responding fiexibility. Each new verb introduces a new series of ideas, and opens a boundless field for practice. Diversity in unity is one of the great laws of nature. Isolated verbs present but vague ideas ; they require to be incorporated in phrases which de- termine their meaning. The learners, after con- jugating a verb with the master, so as to make sure of its pronxmciation, if not already com- pletely known, must, at first, with his aid, then alone, join words to it in all its moods and tenses, attaching thereto, directly and mentally, the ideas conveyed by these combinations. The conjuga- tion of propositions, both useful and interesting, supplies memory with the materials of conversa- tion; whereas the monotonous task of conju- gating the verb by itself presents no distinct idea to the mind, and affords no help toward speak- ing. Hence it is, that the ablest teachers at the' present day exercise their pupils on the verbs in the way we suggest. 116 THE STUDY OF LANGUAGES. Learners, beside fomiing different phrases on tlie same verl), should compose variations on the model phrases, which the professor selects from what he has read to them. This exercise is superior to the first ; because it leads them from the known to the unknown, and presents to them, not words with which to form phrases, but sentences to be decomposed into their elements and recom- posed by imitation. The professor will give the preference to those expressions which elucidate points of grammar, especially those which differ in their construction in the two languages, taking care in these variations to preserve the syntactic or idiomatic character of the model-phrase. If his pupils hesitate, they should be at once assisted either by his recalling to their memory the model-phrase, or stating the grammatical rule which governs it, as also by constructing for them a part or the whole of the sentences proposed. When seasonable assistance is afforded to them, more sentences are formed in a given time, the syntactical construction of the language becomes very familiar, and the words are learned with their proper pronunciation, by being repeatedly heard from the instructor. This exercise is the counterpart of that for THE AKT OF SPEAKING. 117 teacliing how to understand the spoken language as described in the last chapter. The master then formed phraseological combinations in his own language; he now forms them in the lan- guage of the learners, for them to translate into his. But, in the present exercise, the introduc- tion of a great number of new words is not so favorable to progress as the reiterated use of those already known. What is required for the ex- change of thought is not so much the names of things as the power of affirming, denying, and questioning respecting them ; one must especially be able to adapt the foreign phraseology to the requirements of thought, and reproduce it in con- versation with the spontaneousness of the native idiom. The vocabulary of young cHldren is very limited ; and, yet, how readily and fluently they speak ! To exercise the judgment and invention, and to afford learners opportunities of applying what- ever knowledge they acquire, should be the con- /stant endeavor of an instructor. Half the knowl- edge with twice the power of applying it, is better than twice the knowledge with only half , the power of application. The model-phrases must, at first, be very sim- 118 THE STUDY OF LANGUAGES. pie, but always complete and composed of ele- ments familiar to the learners, in order that they may turn their attention exclusively to the con- struction. The master will modify them, by sub- stituting for the words which enter into their composition other words of the same species, changing the moods, tenses, and persons of the verbs, putting an interrogative or negative propo- sition for an affirmative, and vice versa; adding adverbs, conjunctions, and other words as they advance. He can always proportion the diffi- culty to his pupils' proficiency, and give them the explanations they want ; but each new word he introduces must be dwelt upon in order to render its application famihar and fix it in their mem- ory. This practical syntax is thus doubly bene- ficial ; but what makes it still more so in public instruction is, that the sentences proposed to each learner, not having been committed to memory, are so many problems, the solution of which in- terests the whole class. / Every question should be asked before naming the learner who will have to answer it. By this means all the members of a class, in the expecta- tion of being called upon, will be induced to listen attentively, and, under the stimulus of emula- THE AET OF SPEA:KrNG. 119 tion, will call forth their mental powers. Each will thus contribute, by the force of example, to the progress of the others, which is not the case when they repeat a lesson learned by all. In order, with still more effect, to direct the attention of his pupils to what characterizes the genius of the foreign language, in the model- phrases offered for their imitation, the professor will write each of them, or have them written on the blackboard ; then, after making sure that all the members of the class understand it perfectly, he will exercise them in making similar ones, as already explained, adducing rules in , aid of the practice, wheneyer required. As a prelude to this exercise, the professor will, at first, confine his pupils to the bare imi- tation of the text he reads to them. After each phrase which he gives them to translate, as ex- plained in the preceding chapter, he will call on them to reproduce the original from their own version, phrase by phrase. Having the text before him, he could easily correct their eiTors, even if he was not thoroughly master of the language he teaches. This second translation, as it requires a knowledge of the pronunciation, more particularly suits the Latin, in which it may be 120 THE STUDY OF LANGUAGES. had recourse to at every stage of the learner's progress. But the reproduction of a text in a language which is to be spoken, is not sufficient for the re- quirements of conversation. We should have it in our power to vary the form of every sentence. It is hy analogy that model-phrases can be modi- fied so as to reproduce similar ones, and thus to multiply indefinitely the expression of thought. ' The learner will then practise making variations on the text, and vsdll modify every sentence in a hundred ways before he parts with it. The repe- tition of the same forms, thus applied to ever- varying ideas, is both impressive and recreative. f It is in living languages especially that the ex- / ercise in phraseology presents great facilities ; for ' the teacher can always supply from his own re- sources abundant materials. If he is a man of education and a native of the country, the lan- guage of which he teaches, he can generally de- termine, without reference to books, the forms of expression which are admissible, and the precise ideas which, in different cases, are attached to words. The model-phrase being previously known, the exercise here recommended conforms to these THE AET OF SPEAKING. 121 vital principles — the idea tefore the sign, omd ihe< phrase ief&re the words / it possesses also tMa great advantage : wMle the judgment is directed to the idiomatic arrangement of the words, the memory lays hold of them as a necessary consequence of their frequent recurrence in the various modifica- tions which the same sentence undergoes, and lays hold of them in an order conformable to the genius of the language. Jacotot, the originator of " Universal Teach- ing," and other eminent professors after him, ex- ercise their pupils on these variations ; but, for the most part, they confine themselves to one text, and to a very limited number of ideas and expres- sions. l!^evertheless, they are on the right road ; we only suggest a more extensive application of their method, a phraseology more diversified. Analogy, which presides over the formation of phrases, is an ismitation modified by judgment : it produces similar forms ; whereas, imitation re- produces the forms themselves. It is a logical process with its premises and its consequences, a species of rule of three in which the fourth term of a proportion is to be xiiscoverei If, for in- stance, the French of I am, hungry is fai favm, what is the French of he is himgry, you are 122 THE STUDY OF LAlfGITAGES. hiungry, are you hungry, etc. ? Tlie solution of these questions is a logical consequence for any one wlio knows the Frencli verb et/re (to be). With a tboroiigb knowledge of tbe conjxiga- tions and tbe pronunciation, these oral composi- tions may be entered upon *at any period of the study. If regularly practised at short intervals, they will gradually be delivered with greater de- cision and fluency ; and a close association will soon be established between the thought and the phrase, which fixes the latter in the mind and begets the faculty of reproducing it spontaneously, of thinking aloud, as it were, in the foreign lan- guage. Translation, in oral expression, is attended / with the same inconvenience which is attached to it in the first two arts, from the want of identical- ly corresponding terms in the two languages. In colloquial intercourse no time is allowed for the operation ; its tediousness could not but be pain- ful to impatient hearers. Unless the exchange of ideas is direct, there can be no genuine conversa- tion. Direct speaking will be the more effectually accomplished, if the learner, conforming to the order prescribed by reason in the successive acqui- sition of the different departments of a foreign THE AET OB' SPEAKING. 123 language, has previously associated ideas witli the words by direct reading and direct hearing. There is no reading-book which does not pre- sent abundant forms of speech fit to serve as models, and which could not familiarize a learner vidth most of the peculiarities of the language, leading him, by analogous constructions, from the simplest phrase to the most complicated proposi- tion, giving him, in short, the power of modifying the expression according to the requirements of thought, an advantage never to be gained from the dialogues which young people are often com- pelled to commit to memory. A person learns to speak, to extemporize, not by reciting ready-made phrases, but by forming them himself by analogy, and adapting them to the ideas he has to express. The acquisitions of memory are limited ; those of judgment are without bounds. Numerous collections of din.lo gnea a,nd de- tached phrases have been compiled for the use of leame];3 ; but why should their power of expres- sion be restricted to such fragments of conversa- tion ? What need is there for all these books of phrases, when good writers abound in expressions perfectly correct, while their application is ren- dered striking and easy by the very textof which 124 THE STUDY OF LAJSTGTTAGES. they are a part ? Besides, the automatic associa- tion of words, learned in a given order, creates a habit which forbids their being availed of in the diversified circumstances of social intercourse. A judicious selection of phrases may, how- ever, become a powerful auxiliary for the mani- festation of thought, if, instead of learning it by heart and adhering exclusively to the text, the learner be exercised in expressing his own ideas by numerous modifications of its phraseology, which render the foreign idiom familiar to the mind. By learning a phrase, the student exercises his Imemory ; but, by constructing one himself, he ex- ercises his judgment. In the first case he only knows that phrase ; in the second he learns with the phrase the rule by which it is formed. By the first exercise he repeats a lesson ; by the sec- ond he speaks. In learning dialogues one tries to retain foreign phrases corresponding to the na- tive, without having occasion to inquire iijto the genius of either language, while the practice of phrase-making obliges the learner to compare the constructions of the corresponding sentences. The decomposition of model-phrases, in order to reconstruct others of the same Tiind, is a felici- THE AET OF SPEAKING. 125 tous application of the analytical method, which leads from the whole to its parts, and by which / we become thoroughly acquainted with the func- tions of the words, the import of their inflections, and the relations existing between them — con- siderations no less important for accuracy of thought than for propriety of expression. A good collection of dialogues would, howev- er, present this practical advantage, that, abound- ing in phrases of daily use, no time would be lost in search of those which are most wanted in ordinary intercourse. It would forward the learner's progress in familiar conversation, pro- vided he should modify them variously in direct association with the ideas. With a perfect knowl- edge of pronunciation and analogy for a guide, an adult especially can practise by himself phraseolo- gical variations which embody ideas directly. This exercise does not come within the province of a teacher, who can assist his pupils in the construc- tion of foreign sentences only by a reference to the corresponding expressions in their native tongue. In fact, the only departments of a lan- guage for which an earnest learner, with a due share of common sense, really requires the aid of a teacher, are audition and prommciation ; for 126 THE STUDY OF LANGUAGES. ' eyery tliiiig else lie can dispense witli and think for himself. Analogy enables ns to adapt the form of ex- pression to the requirements of the thought ; it is the light of language, its vital principle ; it pre- sides over its formation, and greatly facilitates its intelligibility, use, and acquisition. When custom is doubtful, analogy decides. To establish analo- gies is the first exercise of the judgment : it is the Mnd of reasoning most accessible to all capa- cities. Throughout life we form new phrases from those we have heard or read. Analogy is the very soul of the phraseological exercises we recommend ; those who will resort to them with perseverance will acquire the habit "of speaking in conformity with the genius of the language. The numerous applications we make of this power in acquiring our own render its action so instanta- neous that it passes unperceived; this accounts for its being so generally neglected in the teaching of languages. The learning of grammar, with a view to con- form to the genius of a language, is contrary to the dictates of nature and reason ; since, as was shown, it places precept before example, theory before practice. The learner must study the facts THE ABT OF SPEAKING. '^ 127 ttemselves, not the rules wHcli have been de- duced from them. A foreign language ought un- doubtedly to be known grammatically ; but this does not mean that it should be learned through grammar ; it means that it should be spoken and written conformably to the practice of the best speakers and writers. If we reflect that gram- marians do not impose laws, but only state, within certain limits, what is the common usage among those who speak and write well, it will be obvious that the readiest and most direct way of ascer- taining this usage is to frequent the society of well-educated people and study the best writers. "W"e shall thus learn from them, as the gram- marians themselves have done, what constitutes correct expression. Custom is the arbiter of lan- guage, and, consequently, should be our guide in its acquisition. In speaking or writing a foreign idiom, we ought to be able, as in the vernacular, to ascertain the right pronunciation, orthography, gender, inflection, granamatical concord, and order of words, by an appeal to our consciousness of their correctness, resulting from reiterated impres- sions, rather than to our recollection of rules. The rapidity of speech in extempore speaking does not permit rules of any kind to be applied : 128 THE STUDY OF LANGUAGES. the words should flow in their right order, not by the aid of reasoning, but instantaneously, from a sentiment of analogy, and as the immediate con- sequence of the thought. There is no time for thinking about the details of composition or pro- nunciation in the very act of speaking ; the mind being then preoccupied with the ideas, their con- nection, and subordination. An art is fully known only when practised without reference to its ele- ments or principles. Those, however, who aim at a systematic knowledge of the foreign grammar must, when sufficiently advanced, study its rules and put them in practice by numerous applications, taking as models the examples which accompany them. It is not by learning, but by applying the rules, that grammar is really known. This is an exer- cisQ of the judgment vastly more interesting and beneficial, in every respect, than mnemonic les- sons of grammar. The multiplicity of these ap- plications, by rendering the syntactic forms ha- bitual, will fix the laws of the language indelibly in the mind. E'othing is well known until it is well understood ; and the best way to understand a thing is to put it in practice. If, for example, a great diversity of sentences THE AET OF SPEAKING. 129 be formed witli Frencli verbs requiring either d or de after them, the perplexity which these prepo- sitions present will soon disappear by that prac- tice ; and learners will use them, as it were, in- stinctively by the force of habit and analogy. In the same way, also, a practical acquaintance with the genders of nouns would be easily gained in the same language which attributes to the names of inanimate objects a masculine or feminine, devoid of any distinctive marks by which it can be known, and whose articles and adjectives vary in their terminations to indicate their concord with the substantives preceded by articles and other determinatives, or joined to various adjectives, associations would be formed in the mind of the student, which would enable him to use the proper gender spontaneously and unconsciously, as do the French themselves. In this manner all the gram- matical forms of a language would soon be ren- dered familiar. The same may be said of idiomatic phrases : let a student, for instance, apply the French ex- pression, perdre quelqyUvm, de vue (to lose sight of any one), to a sufficient number of phrases to render the construction habitual, he will not be liable to forget it, and afterward fall into the bar- 130 THE STUDY OF LANGUAGES. barism of a literal translation of tlie correspond- ing Englisli pkrase. The power of expression will arise from the frequency and diversity of these applications. Eepetition is necessary to produce lasting im- pressions on the brain ; hence the power of speak- ing arises from habits engendered by patient prac- tice ; and, as the professor cannot always, in his lessons, devote to the phraseological exercise the time required for imparting to his pupils the fa- cility and variety of expression which thoughts demand, he will give them, or they will them- selves select from a book, the model-phrases on which, in the intervals of the lessons, they will write and make vi/oa voce analogous variations. The phraseological constructions which the student practises by himself, are the fittest to se- cure to him the power of speaking the foreign language. But, to this effect, he must, when ex- ercising himself in the absence of the professor, carefully avoid forming the sentences in his own language for translation into the other. He will express the ideas directly in the latter, that is, will think aloud in it. After a few successful attempts in this, the transition to carrying on mentally a train of thoughts will present no difficulty, espe- THE AET OF SPEAKING. 131 cially to one who can follow directly the ideas of a book or of a speaker. There can be no conversation on a connected subject, tmless the thought be embodied directly in the language. The working of the mind in the act of translating impedes the exchange of ideas in speaking as well as ia listening ; it checks the movements of the heart, under whose inspira- tions are manifested the intonation of the voice and the expression of the countenance, so indis- pensable to the force and clearness of oral dis- course. In order the more certainly to make the for- eign language'the vehicle of his ideas, the stu- dent, speakiag to himself as a child does when talking to her doll, will connect together familiar phrases — affirming and denying, questioning and answering in turns — carrying on, in fact, various trains of ideas as prompted by his imagination: He will, in these monologues, iatroduce the words and phrases he knows best, and which flow nat- urally from his lips with tones and inflections of voice that harmonize with what passes in his mind. The written exercise, which will serve at the same time as a preparation for the art of writing^ 132 THE STUDY 01" LAITGUAGES. should precede the oral exercise, if the organs of hearing and speech have not yet been sufficiently exercised in pronunciation. In any case the two exercises will assist each other. If a person, haying once acquired a language, perseveres in these exercises, conjointly with direct reading, he will preserve — ^nay, more, will in- deiinitely extend his knowledge of it. Habit, not skill in the practice of an art. is the means of never forgetting it. Although, in our ordinary social relations, the exchange of thought is chiefly effected by de- tached phrases, we sometimes have occasion to unfold at some length our ideas, to relate facts and incidents. Passing, therefore, from detached phrases- to connected discourse, and from transla- tion to the direct expression of thought, the learn- ers will acquire the talent of speaking extempore in the foreign language, as they would in their own, by relating anecdotes, historical facts, re- markable events, or other narratives, very short at first, and becoming gradually longer as their pro- ficiency increases. He who has, one day, spoken for ten minutes, has surmounted a difficulty, and will be able, another day, to speak for fifteen minutes. So will this invaluable talent progress. THE AET OF SPEAKING. 133 The student will take the subject-matter of his narrations from works written in the language he is learning; but he must be careftil not to make them mere lessons of memory. To recite is one thing, to speak another. The learner should neyer parrot words in a given order, but should dehver his narrative as an original language of his own, a task which presents no difficulty at this advanced stage. In studying his subject he must, therefore, pay more attention to his ideas than to the manner in which they are conveyed, to the succession of incidents rather than of the words, and afterward relate the story in his own way, availing himself of the author's phraseology only when it comes without effort as the direct ex- pression of his ideas, and even modifying the incidents or substituting others when memory fails. In a class all the learners may prepare the same nan-ative, and each, in turn, deKver some portion of it. Should they be sufficiently ad- vanced to enter upon this exercise, and yet feel diffident, they ought to select subjects which are familiar to their instructor, thereby enabling him to aid their recollection. The professor, by skil- fully catechising them respecting the characters 134: THE STUDY OF LANGUAGES. and incidents of the story, through the mediimi of the foreign language itself, would elicit irom them the whole matter. In an advanced class, considerable interest and benefit would accrue from the introduction by the more forward learn- ers of narratives not previously known to their class-fellows. Narration is in a foreign, as in the native idiom, the best preparation for extemporaneous speaking, and no less useful as a means of intel- lectual culture than necessary in social inter- course. It yields in importance to no other exer- cise. It appeals to the judgment as well as to the memory, by directing attention to the lan- guage of the author, and also to the connected facts of the narrative, to portraits of characters and descriptions of places. It fosters in a class self-confidence and presence . of mind, without which words and ideas are unavailing for public speaking. It possesses also this advantage, as an exercise in pronunciation, that the latter is natu- rally associated with the ideas, whereas, in oral reading, it is irrationally made a consequence of the orthography. The professor will again promote the progress of his pupils in extempore speaking, if he reads or THE ART OF SPEAKING. 135 delivers to them nan-atives, descriptions, or any interesting discoiirses whicli, while exhibiting to them the proper way of treating a subject, will also set them the example of correct diction and good pronimciation. Success depends on the frequency of the efforts made to reproduce what has been read or heard. It is a common saying that, to learn to speak, one should speak much: this is not sufficient; imitation of what is read and heard is far more conducive to improvement in that art than the act itself of speaking. Mere practice in the latter imparts volubility in dealing out one's stock of materials ; it does not enrich the mind with one word, one idea. It has been shown that reading to learners is preferable to conversing with them, as a means of initiating them into the art of hearing ; so is narrating preferable to joining in conversa- tion as a means of learning to speak. It affords greater facilities for the intelligent imitation of good models and the delivery of long discourses ; whereas, the use of monosyllables and detached phrases being optional in conversation, the learner might not make the efforts necessary for long periods. The instructor, owing to his greater 136 THE STUDY OF LANGTTAGES. power of expression, would probably engross the conversation to himself, so that the desired end would not be attained. Even, if the pupil took an equal share in it, he would also speak less in a given time than in narration. The latter presents this other advantage, that the professor, having neither to supply the subject of conversation, nor to answer his pupil, would give his attention exclusively to the manner of his delivery, and might thus correct all his errors. The case would be quite different in conversation. To keep it from flagging, the professor would neg- lect the manner for the matter ; his mind being preoccupied with what he should say, or what re- plies he should give, the pupil's mistakes would pass unnoticed, and might become a habit. iSTo conversation is possible, unless one can speak with sufficient ease and correctness not to give occasion for frequent inten-uptions, caused by hesitation in the choice of words and fonns of speech, or by the con'ection of errors in pronun- ciation or construction. These interruptions di- verting attention fi'om the subject to the words, would be equally discouraging for the pupil and fatiguing for the master : they would render se- rious conversation almost impossible. THE AET OF SPEAKING. 13Y . Besides, what conversation can there be be- tween a master and Hs pupils ? The very little that the latter conld say would never afford suffi- cient practice to gain an extensive range of collo- quial language. They meet, the one to commu- nicate ; the others, to receive instruction ; the for- mer ought to speak ; the latter, to listen. As to those unconnected, commonplace con-' versations which daily arise from the familiar intercourse of hfe, and for which dialogues are a sort of preparation, they do not require any ex- tensive knowledge of a language, nor any remark- able conversational powers. An intelligible pro- nunciation and the most ordinary phraseology are amply sufficient. It is worthy of remark that, in in this familiar exchange of ideas, the ignorant speak more than the learned, and fools more than men of sense. The loquacity of young children and servant-maids is proverbial. Descartes must have alluded to this kind of talk when he said, "Mere speaking does not require much judg- ment." As an introduction to this colloquial talk, learners may be asked in the foreign language itself various questions illustrative of the verb or the model-phrase, which is in course of practice ; 138 THE STUDY OF LANGUAGES. and, by substituting tbe affirmative or negative proposition for tbe interrogative used by the pro- fessor, tbey will find no difficulty in giving the answers in tbe very words and idiomatical or syn- tactical construction of the questions. With a higher aim in view and at a more ad- vanced period, if the professor has sufficient lei- /sure to converse more at length with his pupils, / and has notpresent to his mind a fit subject in ; which the latter can join, we vsdll suggest that every object within reach or within sight may supply matter for instructive conversation. An enlightened teacher can always draw out his pupils by questioning them on the forms, colors, dimensions, and other properties of any article whatever, on its value, origin, mode of fabrication, and on the substances which enter into its com- position. By the relations which this article may bear to others, he will be led to conversing with them on a great number of things, which cannot fail to extend theit vocabulary and enlarge the sphere of their ideas.* Let the professor be received as a friend in the See " Language as a Means of Mental Culture and Interna- tional Communication." Book IV., Chapter II. Conversations on Objects. THE AET OF SPEAKING. 139 family of Ms pupils ; they will then have the op- portunity to put into practice in his society the knowledge which he has imparted to them of his language. In any case, connected conversations can be practised only in social intercourse and in private teaching ; they are next to impossible in large' classes. The members of those must resort to narrations, to which they may add recapitula- tions of their readings. These exercises so very superior to conversation, as means of unfolding the talent of speaking, will, at the same time that they afford an opportunity of turning to advan- tage the acquirements gained by the practice of the first two arts, exercise their fellow-students in the understanding of the spoken language. At this last stage of their studies, they can commit but few mistakes in speaking, and should not, therefore, be deterred by any motive from joining in conversation. Self-confidence is the basis of success in every art. From the moment they use the foreign language with any degree of expertness, their further improvement will be car- ried on, as in their own, by frequent intercourse with the well-educated, the reading of the great writers, and due attention to precepts on the ora- torical art. But it is not to be expected that, in 1 140 THE STTTDT OF LAl^GITA&ES. ordinary circirnistances, a person wiU he able to acquire at liome a complete knowledge of all tlie idioms and delicacies of a foreign language, or command of expression adequate to the elaborate discussion of serious subjects. Great powers of elocution are rare in the native and rarer still in a foreign idiom. The different exercises we bave sketched, by wHcb learners are led from their first initiation in the understanding of the spoken language, in pronunciation, phrase-making, and narration, to the complete possession of these different depart- ments, demand the active cooperation of master ,and pupil. They must alternate at intervals, proportioned in length to the age of the learners. The prolonged action of the faculties on one and the same subject fatigues them and produces in- attention. A change of occupation is indispen- sable to renew their energy. Students will be less subject to fatigue and listlessness, as we have suggested, if they are of an age to take pleasm'e in the work allotted to them, and, especially, if they have gained the talent of direct reading. It is in this case only that the professor can really take an active part in their studies, and that they will derive from THE AKT OF SPEAKING. 141 his services all the advantages to be expected therefrom. The great talent of the master lies in exciting his pupils to study, in leading them to voluntary efforts, and in making them feel the necessity of the tasks imposed on them. When he has once put them in the right track, their progress will thenceforth depend only on their perseverance in following the prescriptions of nature. -The great- est service a professor can render his pupils is to make them independent of him. Some persons lament that they have not a talent for learning languages : this is often hut a plea for a want of energy, of perseverance, hut also because they know not how to proceed in the study. Man has been created a communicative being, and has been endowed with the means of accomplishing his destiny. Every language is accessible to his intellect, and its- pronunciation to his organs ; all that he requires is the will and a good method : a child with a lever is stronger than Hercules abandoned to his own strength. CHAPTEE V. THE AET OF WEITIITG. " Iter est longum per prsecepta, breve et efficax per exempla." Seneca. In the ordinary circumstances of life we have less frequent occasion to write than.^-o speak. The necessity for composition in a foreign lan- guage is almost entirely confined to epistolary- purposes ; and very few are those who, being ac- quainted with a foreign language, have corre- spondents abroad. "With the exception of diplo- matists and merchants, out of a hundred persons who learn foreign languages, not two perhaps ever have need of writing them. Besides, the fear of not expressing one's thoughts clearly, or of maMng blunders, often deters persons from turning this talent to account. Nevertheless, the art of writing, though of limited utilityas an ultimate object, is a power- ful auxiliary to the acquiring of a critical and complete knowledge of a language. The exer- THE AET OF -WltlTING. 143 cises necessary for improvement in tlie art of wri- ting a foreign language, will enable a student, without a master's aid, to become well acquainted with tbe spelling, concord, and arrangement of words, and to ajiply tbe rules of the language to ,tbe expression of tbouglit, as lie learns tbem. TMs art is the more easily acquired, as, like tbat of reading, it can be practised in the master's ab- sence, being based exclusively on the imitation of good writers. The various exercises recommended for learn- ing the art of speaking, are also applicable to that of writing. The student who understands the foreign tongue, both written and spoken, without translating, ought always, when attempting the construction of phrases, either orally or in writing, to attach to them their correlative ideas, without the intervention of the corresponding expressions of his own idiom. In proceeding thus gradually from the simplest to the most complicated propo- sitions, from detached sentences to connected dis- course, he will soon be able to compose directly, that is, to think in writing that language. The greater the proficiency attained in the arts of reading, hearing, and speaking, the greater the success in the exercises which have composi- 144: THE STUDY OF LANGUAGES. tion for their object. This is the order dictated bj reason, but reversed by routine despite com- mon sense. To attempt writing a language with- out previous extensive reading, is wishing to reap without having sown, to know without having learned. The exercises compiled for the avowed object of applying rules at the outset of the study, are purely mechanical, and can end only in dis- gust and failure. We should tax any one with insanity, who should insist on teaching a child to write his own language before he had learned to read ; is it not far more unreasonable and absurd to compel a learner to write in a language of which he has not yet the least practical knowledge? Imitation being the basis of progress in this ' art, recourse must not be had, in the first instance, to grammar, which only gives precepts, but to a good author who may be used as a model. His text, while exhibiting the rule embodied in the example, presents the words with their orthog- i-aphy and their true meanings ; it teaches, in ad- dition to syntax, all that constitutes the merit of I style. It also dispenses, and this is not one of its least advantages, with the use of the dictionary, which is still more perplexing in writing a foreign language, than in translating into one's own. THE AET OF ■WAITING. 146 It was by studying tlie works of their pred- ecessors that the most , distinguished authors learned to write. Many of them have declared the fact in their works, and, eager to benefit us by their experience, have earnestly recommended the practice of reading. " Plato," says Longinus, who himself holds the same opinion, " has taught us that the surest means of attaining perfection in style, is to imitate and emulate eminent writers." D'Alembert exclaims, " "What precepts are pref- erable to the study of great models ? " — " The as- siduous reading of good writings," says Yoltaire, " will be more useful for the formation of a pure and correct style, than the study of our grammars. We soon acquire the habit of speaking well by the frequent reading of those who have written well." One of our most eloquent writers, J. J. Rousseau, also says : " I give you no other rules/ for writing well, than the books which are well written It is a law of nature that our minds insensibly imbibe a coloriag from those with whom we associate, whether they are brought in contact by the living voice or the written page." Demosthenes, in order to improve his style, tran- scribed eight times the " Peloponnesian "War " of Thucydides. "When Lord Clarendon was en- 146 THE STUDY OF LAWGTTAGES. gaged in writing Ms history, lie was constantly studying Livy and Tacitus. The latter classic was also the favorite author of Montesquieu. Again, Benjanoin Franklin, adopting Dr. John- son's opinion, made the Spectator of Addison his model-book. Byron acknowledged what he owed to Pope, when he mentioned him, in one of his letters, as "the delight of his boyhood and the study of his manhood." Boileau declares himself the imitator of Horace. Dante took Yirgil for his model : " Thou art my master and my author," he exclaims in his sublime poem ; " it is from thee alone I took that beautiful style which has done me honor." The best mode of imitation in foreign com- position is douhle-trcmslation, which consists in translating the foreign text into the national idiom, and then endeavoring to reproduce that text by translating the version back into the original. The vwa voce double-translation, recommend- ed in the preceding chapter, was made phrase by phrase, and consequently only taught how to ex- press detached ideas ; the present exercise, having for its object the formation of style, is performed on passages increasing in length with the learner's THE AET OF WEITING. 14:7 progress, and will lead to the writing of connected discourse. This imitative process can be easily adapted to all degrees of proficiency. Besides exercising the learner in the composition of his own, as well as of the foreign language, it suppKes the means of correcting mistakes in the latter, by comparing his second translation with the author's text, which, if judiciously chosen, is the safest guide that can be followed. Double-translation is not an innovation ; it is recommended by Cicero, Pliny the Younger, Quintilian, and nearly all those who, to the pres- ent day, have suggested means for acquiring the arts of writing and speaking in a second language. It is chiefly by means of this exercise that Queen Elizabeth, as Eoger Ascham, her tutor, tells us, learned the Latin language, which she spoke so freely. The historian Gibbon declares that he gained, through double-translation, an extensive knowledge of Latin and French, and the com- mand of a correct style in English. Its principle is generally recognized in classi- cal instruction ; the Latin exercises given to boys, are usually imitations of the classical texts, which they have already construed. Its utility is not 148 THE STUDY OF LAJIGUAGES. restricted to tlie idioms of ancient times, and of modern Europe ; Sir William Jones asserts that, " by double-translation more Arabic and Persian will be learned in ten months, than can be learned in ten years by any other method." The difficulty of the task mnst always be pro- portioned to the proficiency of the pupils ; no ex- ercise ought to be either so difficult as to discour- age their efforts, or so easy as to require no effort at all. Double-translation is perhaps the best of all exercises for avoiding these extremes. It also possesses the invaluable advantage of being not less available in the first than in the last stage of study; for its difficulties may be diminished or increased at pleasure ; and hence its perfect adap- tation to public instruction, as the same tast may be given in common to persons of very unequal proficiency who happen to be in the same class. The choice of the model, especially in living languages, may always be suited to the capaci- ty, progress, or requirements of the learner. In the commencement, the first version is made as literal as the genius of the national language permits; and the second translation is written shortly after the first, while the language of the original text is still vivid in his mind. A perusal THE AET OF -WEirESTG. 14:9 of the text just before making the second transla- tion, would also, if required, facilitate the repro- duction. Moreover, the student can, at any moment, have recourse to it for the purpose of making sure of the words or construction which/^ have escaped his memory. This reference to the model is, in many respects, preferable to using a dictionary. ' As the student advances, he will, in his efforts to reproduce the text, rely more on his knowl- edge of the language and less on the recollection of the original. His first version, too, will be less and less literal; and he will gradually in- crease the interval between the two translations, ) thus leaving greater scope to memory and reflec- ' tion. On the other hand, if he perseveres in the practice of the three other branches, simultane- ously tvith this exercise, he will daily acquire greater facility in its execution. In his first version the learner must be faithful to the original text, and, at the same time, con- form to the genius of his own idiom : he should be careful neither to add to nor take from the ideas ; for his business is rather to copy than to compose. It is often suggested, as a general di- rection, that the foreign work should be rendered 150 THE STCTDT OF LANGUAGES. in that style which, it may be presumed, its au- thor would hare employed had he written in the language into which the translation is made. This suggestion, although consistent with reason, must be received with some caution ; for, if strict- ly followed, it may sometimes lead to mere imita- tion rather than to faithful interpretation of the original ; besides, it may restrain the flexibility of the language and its adaptation to diversified ex- pi'ession of thought. The correction of tliis translation has, in pub- lic instruction, great advantage over that of ori- ginal essays : it can be effected simultaneously for aE the members of a large class ; and it occupies comparatively little time. Each learner havings his version before him and pencil in hand, is, in turn, called upon to read a portion of the compo- sition, on which the professor comments; the others, at the same time, mar!king such mistakes as they have made in common with the reader. The examination of an exercise of twenty lines, read aloud two or thi-ee times over, in portions of four or five lines, by a dozen learners, would am- ply srtffice to elicit all the errors which may have been committed by fifty or more students. Es- says, on the conti-ary, require each to be read V THE AUT OF WEITESTG. 151 tlirougli and examined separately; their correc- tion, consequently, engaging tlie attention of only one learner at a time, leaves the others idle, and, from the time consumed, is impracticable in a class. Translation, by requiring all the learners to express the same ideas, brings their powers of composition to a closer and, hence, a more inter- esting cont'est ; much useful information is also elicited by a critical examination of their different modes of expressing the same thoughts. If the learner, haying no competent person to whom he can submit his first version, wishes, nevertheless, to know whether it is correct, he may compare it with a good translation of his model. He might even make his second version from such a translation, if, from want of time, or any other reason, he wished to dispense with vn-it- ing the first. This mode of proceeding is recom- mended by M. Guizot: " Take," he says, " a page translated from a good author into your own lan- guage or any other that you know ; render this page in the language of the author, and compare your work with the original. By so doing, you learn the words, the syntax ; and enter into the spirit of a language which is fixed in the memory by reading and writing." From this it is obvious 152 THE STUDY OF LANGUAGES. that translations may always be turned to account, whetlier they accompany the original text or are published apart. The second translation naturally presents more diiSculty than the first, and cannot be made with- out the student constantly appealing to his recol- lection of the original text. These very efforts will have the effect of fixing in the memory its words and peculiar construction. The more idio- matic the foreign expressions, the greater must have been the effort to render them accurately, and the better will they be remembered. ■J Let it not be objected that learners would be 'apt to copy the original text, instead of perform- ing the second translation. Such a practice is im- probable at this advanced stage of the study and at the age we have assigned to them, especially if they are not unreasonably rebuked for the er- rors they commit. It rarely happens that young men betray the confidence placed in them. Under the worst circumstances, the second translation may be written in the presence of the instructor ; or the text-book may be taken from them, when they are writing the second translation. For greater convenience, the second version should be written in the same copybook opposite THE AUT OF WEITING. 153 to the first, the alternate pages having been left blank for this purpose ; thus comparison will be considerably facilitated. The assistance of an instructor may be dis- pensed with for either the first or the second ver- sion, if the student has become famihar with the foreign language, and if, more especially, he can write his own- correctly. The better he knows the latter the easier will he fiad this first opera- tion. Even in case of his having followed pur advice, and exercised himself more in direct read- ing than in translating, he will have no hesitation in rendering the ideas of the foreign text ; for a person who can express his thoughts readily in the language into which he translates, will not re- quire much previous practice in translation to suc- ceed in this version. When he clearly conceives an idea, firqm whatever source derived, he will surely be able to express it. In order to correct his second version, the stu- dent wiU compare it vsdth the original, phrase for phrase, word for word, noting the differences he finds between the two ; whether in orthogra- phy or phraseology, he will easily discover the reason of the preference to be given to the expressions of the original text, if he has read 154 THE STUDY OF LAKGUAQES. mucli, and bestowed some attention on gram- mar. In case a student is as yet too young or too little versed in Ms own language to be able to ac- count for the difference of words or construction, he will write the words of the model over his own, and will submit both to his instructor, who wiU find occasion in this parallel for entering into all the developments of literary criticism adapted to his age and proficiency. According to the nature of the questions, the instructor will direct his pupil's attention to the orthography, the etymology, or the grammar ; will point out the various shades of meaning in synonymes; and explain the difference between idiomatic and syntactic forms, between the literal J and the figurative senses, between absolute and / relative terms, between expressions of an elevated and those of a familiar style, as well as the vari- ous modes of expressing the same ideas, or the difference of idea resulting from difference of construction. The doubts that each pupil of a class thus submits to the professor, will afford him an oppor- tunity for giving instruction of the highest value to the whole class, and the more valuable as it is THE AJET OF WEmNG. 155 not to be found in tlie majority of the books usually put into the hands of young people. It will sometimes happen that the student competes successfully with his model; for the same idea may be presented in different ways, and an author does not always choose the best. These httle literary triumphs will greatly enhance the pleasm-e of the compositions and stimulate to perseverance.* The moment when these comparisons are made is not the best time for the student to correct his exercise ; he merely marks the errors with a pen- cil, reserving them for futxu-e correction fi'om memory. The reflection required on a second consideration of the same subject to remember the differences previously noticed, will strengthen the memory and prevent the recm-rence of the same eiTOi*s. Eeflecting on one's faults is the sm-est way of avoiding them in futm-e. K a person who learns a foreign language without a master, has recourse, for the double- translation, to a book with the two texts on oppo- * It must not be supposed that we here flatter students at the expense of authors. The Vicar of Wakefield, for instance, which is frequently put into the hands of French pupils who are learn- ing English, abounds with incorrect expressions. 156 THE STDDT OF LAJSTGTJAGES. site pages, lie may, if necessary, compare his two rersions with tliem ; and if, for want of time, or any other reason, he cannot write the exercise, he will go through it orally, endeavoring to repro- duce each text of the book, without looking at it first ; and, as he advances with his version, he wiU compare it with the model text. The choice of the text to be used as a model for double-translation, depends altogether on the kind of composition and style that is best suited to the learner's ultimate purpose. "What most ^people want, however, is to be able to write let- ters, and this end can be best attained by the use of the class of books which we have recommend- ed for initiating students in the first three arts. The familiarity of their subjects and the simplici- t}?- of their style are perfectly suited to the re- quirements of epistolary correspondence. It would be waste of time to attempt to rival the great writers. I If, when able to write well in prose, a student, /endowed with a talent for poetry, should wish to I try his hand at this sort of composition, or merely i; to study its mechanism, so as to better appreciate its merit, he will again have recourse to double- translation, taking for his model among the best THK ART OF VTEinSG. 157 poets, tJie one whojo -vvritiiig-s are the most eon- gvinal to his taste. The passaicos of poetiy, read by the pT'otessov. as we haA-e already suggested, having formed the student's ear to the rin-tlim and voi-silieation peeiiliar to the language which he studies, he mil find little difficulty in reproducing then\, if he is endowed with poetical discrimina- tion. The first version in pix^se. which affords liim the oeoasion for transferring into his style some of the poetie;il he;nities of his model, wiU not Ml to extejid his powere of composition in Ms own langnage. while the efforts lie will make to repro- duce in his second version the figuratiTe langnage and the harmony of the style, will render him sensible of the difference hetween the literal and the figuratire sense, between familiar and dera- ted expressions, between the style of prose and that of poetry. His TOcabulary wiU also be en- richetl, owing to his being forced, by the neces- sity of rhyme, cadence, and measure, to examine many words which liiffer in termination, quanti- ty, aecenr, or number of syllables, thoitgh nearly of the same meaning. After havini: practised for some time on the ideas of ff<.x>d writers, endeavoring to imitate 158 THE STUDY OF LANGTTAGES. their forms of expression, the student may, with some chance of success, make an attempt at origi- nal composition. Erom the direct expression of thought by detached phrases, the transition is easy to a connected discourse. The words, hav- ing once become the direct and habitual signs of thought, will flow naturally from the pen, if the student exercises himself in the art of writing, as in the art of speaking, in treating subjects with which he has become famihar by reading and hearing. In general,- all interesting and well-written narratives, that have been read with this object jin view, in the foreign language itself, will pro- ' mote the student's progress by the impression which the direct reading will leave on the mem- ory: both the form and the substance are thus equally retained. This practice in composition should also, at first, be combined with corre- sponding oral exercises, in such a manner that they mutually assist each other, either by the pupil's writing the narratives already made viva voce, or by relating to the professor those which have been written. In proportion as the learner advances in com- position, its difiiculties should, be increased by THE AET OF WEITmG. 159 longer pieces, with a greater interval between tlie reading and the writing, until, at last, he can write without any preparation. For one who has stored his mind with various knowledge and cultivated his taste by the assiduous study of standard authors, the moment has come to rely solely on his own capabilities. He stands in pre- cisely the same position with regard to the foreign language as to his own ; and his improvement in both vrill be secured by the same means. He will devote his attention, according to circum- stances, or the peculiar turn of his mind, to the epistolary style, to the relation of facts that have come to his kaowledge, to the descriptions of ob- jects, places, or persons, to summaries or analyses of his readings, or to any other subject on which he has clear and accurate notions. "When reading, hearing, phrase-makiag, and oral and written narration have, by the reiterated association of ideas with their signs, imparted the habit of thinking in the foreign tongue, as we think in our own, the one will not be more liable to be forgotten than the other. Both languages, equally making one with the thought, wiU adhere with the same tenacity to our individuality, and win be permanently assimilated with the elements 160 THE STUDY OF LANGTTAGES. of our intellectual constitution. The organs themselves through which this point is gained, the eye and ear, the instruments of curiosity, the tongue and hand, the instruments of imitation, contract habits corresponding to those of the faculties which direct their action. The habits of language thus acquired will be infinitely more correct than those which an adult derives from a residence abroad. The student who follows our method, having only good models before him, either in his books or in his master, runs less risk of being misled than a person in a country the language of which he is yet unable to speak. Indeed, in this last case, being fre- ■ quently thrown into the company of people who speak incorrectly, and being under the necessity of expressing his wants, of communicating his sentiments, before he has heard the words and ' phrases often enough to be able to reproduce them accurately, a person must make mistakes which, being seldom corrected, become habits never to be got rid of. As a compensation, how- ever, he gains in facility what he loses in correct- ness. "Without unfolding the means by which a correct style is gained, as it is the very same THE AKT OF TVErnXG. 161 which is required to attain a similar end in his own language, we wiU repeat onr previous assertion that the art of writing well is the fruit of assidu- ous study of great writers, and of continuous efforts to imitate them, merely ohserving that it win he sufficient, in ordinary cases, to take for models texts in a simple and familiar style, as being the best suited for conversation and letter- writing. Beyond this kind of composition we do not see any necessity for striving after a high degree of perfection in writing any language but one's own. The most eminent Latinists are not those who write their own language best : and never has the Latin versification of the colleges made a good T'-Tigbsb poet. The marked difference which characterizes the genius of languages wiU scai-ce- ly allow any one to write two in perfection. Such is the force of habit that if we express our thoughts in a foreign idiom, to the exdusion of our own, the forms peculiar to it vdll at times find their way into the latter, both in speaking and in writ- ing. It is therefore advisable to restrict rather than to encourage composition in a foreign idiom. In support of this truth we will quote Vol- taire, who speaks from experience : " On my re- 162 THE STUDY OF LANGUAGES. turn from England," he says in a letter to Lord Bolingbroke, "when I had passed nearly two years in constant study of your language, I found myself embarrassed while composing a French tragedy. I had almost accustomed myself to think in English : I felt that the words of my own language no longer presented themselves to my imagination with the same abundance as be- fore : it was like a stream, whose source had been turned aside ; much time and effort were required to make it flow again in its original bed." Grammar, rhetoric, and logic, will find their application as the complement of linguistic stud- ies. To possess a critical acquaintance with a language, we must have studied its genius, have become thoroughly imbued with the principles which form its groundwork, and have analyzed expression in its relations with thought. The exercises above recommended having their source in two powerful instincts, curiosity and imitation, and being, as reason prescribes, the actual practice of the arts to be acquired, success is infallible. They possess another great advan- tage, they are admirably adapted for public in- struction, inasmuch as they afford to students of different degrees of proficiency the means of de- THE ART OF WEITING. 163 riving equal benefit from the master's lessons, ■without loss of time for any of them. This ad- 1 vantage is especially owing to the fact that, as the j pupils never perform in his presence the work '■ they can do by themselves, he thus has time to ' teach them what they could not learn without his aid. Our method thus assigns to both pupils and master, their proper sphere of action. The for- mer learn the written language, the latter teaches the spoken language. This method dispenses with all the prepara-i tory lessons, which only delay the, practice of the/ language. Eeading-books suffice for all the re- quirements of study. More than one volume at a time is never required for practising the differ- ent exercises it prescribes ; and this is not one of its least advantages. A page, selected at random, in the first volume that comes to hand, offers the professor all the facilities for exercising his pupils in the practice of the four arts, for teaching them orthography, pronunciation, grammar, in short, every thing that relates to composition in the- two languages. It renders the learning of a language accessi- ble to adults as well as to children, to the poor as well as to the rich, by supplying means of self-in- 164 THE STUDY OF LANGUAGES. struction, and requiring the help of a master only for the second half of the language, when the learner already knows the first and most impor- tant half. Many persons actively engaged in business feel the want of knowing a foreign language, who are deterred from the attempt to learn it, on account of the tedious labor of the dictionary, and of all the drudgery imposed by routine as an indispensable preliminary. If the student, after spending much time on these unprofitable tasks, is compelled to discontinue them, he finds that he has learned nothing really useful, and retains only the painful recollection of the labor and fatigue he has gone through. By following simple and natural processes in harmony with the end proposed, such as those we recommend, there can be no doubt that, in the maturity of reason, even at an advanced age, a person might, in six months, acquire what is use- ful in a living language, better than a boy of ten could do so in as many years, by the ordinary rou- tine. The greatest linguists, from the Scaligers to Elihu Burritt, the learned blacksmith of Massa- chusetts, who is said to have learned above twenty languages, have nearly all acquii-ed them in the THE AET OF -witrmsrG. 165 maturity of life, and without masters, by follow- ing a method similar to the one we have sketched. Plutarch, who began the study of Latin late in life, made rapid progress, because, as he him- self says, his knowledge of things enabled him to enter into the thought of the writers. Themisto- cles, also advanced in years, learned Persian so well in one year, says his biographer, that he used to converse with the King of Persia on state affairs better than the Persians themselves. Cato the Censor learned Greek in his old age, and knew it thoroughly. Alfieri began the study of that language at forty-eight, and attained a high reputation as a Hellenist. Sir William Jones had passed his thirtieth year when he began to learn Eastern languages, in which he is known to have been deeply versed. Ogilby, the Enghsh transla- tor of Yirgil and Homer, had been a dancing- master; he did not know a word of Latin at forty, nor of Greek at fifty-four. Matigard, a distinguished man of letters, became, after three months of study, a successful teacher of Itahan and Spanish, which he had learned in his sixtieth year- The celebrated Dr. Johnson undertook, when seventy years of age, the study of Dutch, with a view to test his capability to learn : the 166 THE STTJDT OF LAITGITAGES. success of the experiment fully satisfied Mm that the powers of his mind were still unimpaired. Eichard Cumberland, Bishop of Peterhorough, at the age of eighty-three, learned the Coptic lan- guage, in order to read the Coptic iN^ew Testa- ment, which Dr. "Wilkins had just published. It is fortunate for the perfectibility of man that the practical knowledge of languages is so easily acquired ; for one of the principles of his nature, sociability, impels him instinctively to enter into communion with his fellow-creatures by the free manifestation of his thoughts. What- ever thwarts this instinct is an obstacle to his improvement. Humanity is endowed with capacities which can be perfected only by the combination of minds : the whole mass is animated with a life which lies dormant in the isolated individual. A powerful impulse will therefore be given to the progress of civilization, by introducing into our schools and colleges a better system of linguistic teaching, which will bring nations into intellectual inter- course. • CHAPTEE YI. OH" MENTAL CULTUEE. " L'etude des Ungues est beaucoup plus favorable aux progrSs des facultes dans I'enfance que celle des math^matiques." Madame de Stael. " The study of languages is muoh more favorable to the prog- ress of the faculties in childhood than that of mathematics." These are two distinct categories of students, those who learn a language for purely practical purposes, and those who learn it as a branch of education, either for the development of their mind, or as a means of extending their knowl- edge of the na;tional idiom. To the students of the first category, those whom an intellect, already ripe, inclines to a vol- untary study, the practical method, as explained in the preceding pages, especially commends it- self. It is more especially profitable to those who, having completed their academical education, are preparing to enibrace professions for which the 168 THE STUDY OF LANGUAGES. knowledge of a foreign language is indispensable, and for those, whatever be their age, who, from taste or necessity, wish to enter on new fields of study, or to enlarge the circle of their social rela- tions ; but who, engaged in avocations which leave them little leisure, are averse to collateral studies, of which they do not see the necessity. As regards the second category, including the youth of our colleges, more especially children under twelve or thirteen, the comjpa/ratvoe methodr- offers them all the means of attaining the end toward which their instructor ought to direct their attention, and which requires, on his part, a thorough knowledge of their language, as well as of the foreign idiom. By oral instruction which bears on all the parts of the study, he will take an active part in all the exercises, which will become, under his direction, a real course of intellectual gymnastics. This is, for children, the most important part of scholastic education. They have abundance of time before them, and have no valid reasons for hurrying in the acquisition of a second language ; for they would probably forget it, before they had occasion to use it. The practical method, as it requires little men- OK MENTAL CULTUEE. 169 tal eifort, leads rapidly and exclusively to the mastery of a language ; tlie comparative process, on tlie contrary, by presenting difficulties vsrhicli unceasingly call the reflective powers into action, inures the learners to self-direction and intellec- tual labor, v^hich constitute its chief merit as an instrument of moral and mental discipline, at the same time that it promotes advancement in the national language. The first is best suited for modern languages, the second for the ancient. The noble pages of history, eloquence, and poetry, which the dead languages exhibit, though few in number, will always stand as models of excellence. The beauties with which they abound cultivate and purify the taste, while reflection finds ample exercise in the consideration of thoughts and facts relating to an order of things above the homely realities of ordinary life. Placed beyond the influence of caprice, the dead languages, so long as they are accepted as the groundwork of scholastic studies, and the test of excellence in literary composition, will tend to check the constant fluctuation of living languages. The Greek and Latin classics are in literature, what the works of the old masters are 170 THE STUDY OF LAJSTGTTAQES. in painting. The love of novelty may for a time draw modern nations from the trne principles of taste ; the study of the immortal monuments of antiquity will always bring them back to the true standard. Ancient languages must continue to occupy a large share of attention in the intellectual educa- tion of boys who are destined to pursuits which depend on literary acquirements. It is a nar- row view to consider them as useful only to the learned professions. Acquaintance with them is beneficial not only to the clergyman, the physi- cian, and the lawyer, but also to the archaeologist, the philosopher, the man of letters, and the states- man, for they are the interpreters of ancient monuments, the original receptacles of our laws, the source of our modern dialects, forming a bond which unites European nations with one another, and with antiquity. Their study in- volves that intellectual discipline which gives the greatest possible development to the faculties of man, and is the common ground on which the noblest intellects are brought into contact. Mathematics, far from being, as commonly believed, the best logical exercise, would, if studied exclusively, rather tend to disqualify the OS MENTAL CULTUEE. 171 mind for general reasoning. They confine tlie student to a narrower circle of mental exercises -than languages and philosophy; they habituate him to a routine of argumentation which presents little variety; they awaken his judgment to the relation of quantity, but, neglecting quality and all other important relations, they leave in abeyance the powers of the understanding which are most useful in the ordinary circumstances of hfe. In the study of languages the mind is engaged as in the world; it is formed to all modes of reasoning, to all kinds of argumentation, by the reading and hearing of serious subjects. The reading of good books is a practical logic in the exercises intended to teach speaking and writing; we find the same dealings with words and ideas as in social intercom-se, the same caution and discrimination between rules and exceptions, the^ same exercises of conception, imitation, and in- vention ; finally, the same methods of induction, analogy, and analysis. At every step in mathematical demonstrations there is a constant perspicuity, a straight and limited path marked out, from which it is almost impossible to wander. But in the expression of 173 THE STUDY OF LANGUAGES. thougM, and in literary investigations, the learner has to feel Ms way/ reflect, compare, judge, apply his own experience, weigh probabilities, disen- tangle networks of inconsistencies, and lay bare sophistical plausibihties. In this necessity for a diversified and complicated action of the reason- ing powers consists the chief value of classical and hterary studies. The learning of languages embraces thought and its expression. The operations of the mind may indeed be said to be identical with the use of language. The various acquirements which constitute the complete possession of a foreign idiom- afford, through the exercises indispensable for their attainment, the means of cultivating at- tention and raising the intellectual powers from their original state to the highest degree of im- provement. The mental discipline generated by a rational method begins with those mysterious lessons by which the learner associates signs with ideas, and continues through the whole course of the study, by means of critical explanations, translation, reading of foreign works, and analysis of their style. The disclosure of the thoughts and sentiments of good writers will gratify his curi- osity, excite his sympathies, improve his taste, ON MENTAL CTTLTUEE. 1Y3 invigorate Hs judgment, enrich his memory, and enlarge his understanding. The continual comparison of two idioms which results from one being acquired through the other, keeps observation and reflection con- stantly on the alert. But, as we can compare only such things as we know, the comparative method will not produce all its fruits, unless we possess the art of reading the foreign language. Then only shall we be able to establish, vrith re- gard to thought and style, an interesting and profitable parallel between the native and the foreign writers. We therefore cannot acquire that art too soon in the ancient as well as in the modem languages. To advance safely in reading, children as yet incapable of self-direction, should be assisted in all the investigations to which translation leads. They wUl be told the true meaning of idioms, the different acceptations of words, and the difference of import between those called synonymes. They will be assisted in rendering faithfully the origi- nal text, and in rendering it conformably to the genius of their own tongue. By exercising them in comparing this text with the interpretation which accompanies it, the professor will clear up 174r THE STUDY OF LAJSGUAGES. their doubts, explain irregularities, and enable them to overcome all the difficulties that are presented by the difference of construction in the two languages. As the dead languages are no longer used for the oral exchange of thought, the speaking ex- ercises will be set aside, which will allow more time to be devoted to the explanation and trans- lation of authors. This branch of the study as- sumes ' an iniportance which it cannot have in living languages, and which will prove all the more profitable to the pupils, in proportion to the skill of their instructor in their native tongue, and to tiie extent and soundness of his informa- tion. If translation is objectionable as a means of understanding the foreign text, it is otherwise, when considered as an exercise, either in oral or written composition : it then becomes a powerful auxiliary for improving a learner in his native idiom, and for exercising him in improvisation, from the moment he possesses the art of reading the foreign language. He thus increases his native vocabulary by the use of words which were not previously familiar to him. Aided by careful study of national writers and ON MENTAL CULTUEE. lYS orators, oral translatioa at siglit, if continued for some time, will prove a better preparation than rules for acquiring that magic power — extempore speaking — which instantaneously calls up the most appropriate terms, and suits the form of expression to the idea. Written translation is preferable to oral for forming a good style, and acquiring great powers of expression in the national language ; for the necessity of reading on, in order to grasp the subject, does not allow time to polish our phrase- ology, or to seek the forms of speech which best conform to the genius of our language. In the first version of the double-translation, we practise the art of writing under favorable conditions; because we can then bestow on the work all the reflection required for the choice of words, their idiomatic aiTangements, and the logical connec- tion of ideas. "It is by translating," says De Gerando, " that young people learn best all the laws of the art of writing." D'^embert also says : " If you wish to be one day translated, begin yourself by translating. The work of translation will yield a rich harvest of principles and ideas, and prove an excellent school in the art of writing." . 176 THE STUDY OF LANQUAOES. He who attempts composition witlio\it first laying in a Lu'ge provision of knowledge, will at best deal out none but coninionplaco ideas, and conceal poverty of thought under pom]) of ])lirase- ology. But a second language presents an in- exhaustible Source of interesting eonijiosilions, which, while they servo as models for the manner of treating a subject, afford by translation the best means of practising the art of writing. In translating from a standiU'd work a learner habituates himself to express sound ideas : he is thereby led to reflect on subjects A\'ldch he Inul not previously considered, and his sphere of thought is enlarged, as is also his power ol" ex- pression ; for it is more difficult to render the ideas of others than one's own. In original essays a learner is not always completely master of his subject, and is apt sometimes to modify tlio ideas as they arise in the mind to siiit them to his scanty stock of words. Translat,ion, by binding him to the particular ideas of the author, teaches him to overcome difficulties; original composi- tion, by leaving him the option both of the snb- stauce and the form, only teaches him to a\'oioii in English, In ordinaiy circumstances, the da^cal pro- fessor, who is nece^arilj much better acquainted wirli the language of his pupils, which is his ovrri. tL;m with the Latin and Greek which he teaches. can make the latter sufeervient to their improve- 180 THE STUDY OF LANGITAGES. ment in the vernacular more effectually than could be done through living idioms by foreign- ers, who are seldom well versed in that language. A great difference, therefore, arises in the mode of proceeding: the study of the ancient lan- guages is essentially a means of extending one's knowledge of his own language ; but since little assistance can be obtained from foreign teachers toward any considerable advancement in the latter, an extensive acquaintance with it becomes indispensable as a groundwork for mastering a foreign living language. The ends to be attained by studying these two categories of languages differ essentially : the dead languages are learned for the sake of the national idiom, the living lan- guages for their own sake. These, as necessary vehicles of thought, require the ideas to be direct- ly associated with the words, a process unsuited to the dead languages, which prove beneficial to learners only by being studied through the native tongue. The practical and the comparative' methods cannot be followed simultaneously. The adoption of one implies the rejection of the other. , Those innovators accordingly commit an egre- gious blunder, who propose to substitute the modern for the ancient languages in the instruc- OH MENTAL CtTLTHEE. 181 tion of youth. Although. Enghsh, French, Italian, and German, may faii'ly compete with the latter in force, lucidity, and gracefulness of expression, at the same time that they far sui-pass them in the number and importance of the benefits which they confer through life on their votaries, still they can never supersede them. But, were such a change to take place, foreigners, who would consequently become the most competent profes- sors in what would then be the chief department of collegiate instruction and the best judges of literary merit, must be placed at the head of academical establishments, and invested with the highest university honors. This would be a complete anomaly, a state of things altogether incompatible with the existing form of literary institutions and repugnant to national feelings. The practice of writing the dead languages should be indulged in very sparingly in public instruction, the more so as it consumes a consider- able portion of time, and thus unreasonably and unprofitably lengthens the period of classical studies. The attention of learners ought to be confined to what is really useful in these studies, namely, the reading and analysis of the great writers of antiquity. It is by reflecting on their 182 THE STUDY OF LANGUAGES. ttougMs and their style ; it is especially by trans- ferring, tliro-ugli translation or imitation, their beauties into the national idiom, and not by cari- caturing them in their own, that classical instruc- tion may be productive of real advantage, that the understanding may be exercised, and a com- mand of the native tongue secured. Depth of learning in ancient literature is far from being the test of excellence in the national language. In a rational course of instruction the national classics should be studied as much as those of antiquity. To know Latin and Greek is a great intellectual luxury, but to know one's own , language is an intellectual necessity. In the instruction of youth the national tongue should hold a preeminence, which, until now, has been denied "to it ; for it is more particularly the instrument of the mind's operations, the record of its stores, the manifestation of our feelings, our affections, our intellectuality. Its writers and orators, its genius and resources, - should, among a people careful of their own dignity, occupy young persons from the earliest to the latest period of scholastic instruction. It is in that tongue that they should be taught to think, to speak, and to write. To it belongs by right the OK MENTAL CULTURE. 183- prize of excellence adjudged by the old universi- ties to tlie Latin tongue. It should be considered more honorable, as it is more consistent and more useful, to speak and write English like the best English and American speakers and writers, than Latin like Cicero and Tacitus. In the private affairs of life, as in political or \ international questions, he who speaks or writes the best will always gain an ascendency over his fellow-citizens. Speech is power. The great end of classical and literary education ought to be to ' confer this power, the most useful, the most de- lightful, the most admirable of human acquire- ments. An enlightened American teaching a foreign living language^-^which he only knows how to read, would stiU have a noble task before him, if he endeavored, through translation, to improve his pupils in their own idiom. He would be able, better than a foreigner, to ascertain, by their manner of translating, whether they really understand their author. Direct reading and hearing will, also, like translating, aid in cultivating in youth the two- fold talent of speaking and composing in the national language, if the professor makes them 184 THE STUDY OF LANGUAGES. repeat vivd voce or in writing what they have read or heard in the foreign idiom. This exer- cise will not only test their diligence, but will also, under his direction, he the means of ac- quiring a ready delivery and great power of expression. On the other hand, under the ex- pectation that they will have to give an account of what they read or hear, they will read and listen more carefully, and will thus more firmly impress on their minds the subjects brought under their notice. Narration exercises the intellectual memory, /that which proceeds from understanding the sub- j ject and rests on the connection of the ideas, on 1 the relation of cause and effect, and on that of premises and consequences. This noble faculty plays a far higher part in our mental organiza- tion than mechanical memory, or recollection of words by their accidental collocation, which, having nothing to do with the judgment, con- sists in retaining and repeating words in a given order, rather than in recalling ideas by their logical concatenation. In written as well as in oral compositions founded on imitation, the powers of observation, reflection, and judgment, are constantly brought ON MKVTAL CULTUEE. 185 into activity by the necessity of studying models, comparing them -with their copies, and discrimi- nating between different forms of expression and the different "n^ays of treating a subject. In original compositions, those of a pxirely naiTative character, resting on the chain of inci- dents, exercise more especially memory and im- agination ; -vrhereas, descriptions and dissertations, without rejecting the aid of these two faculties, caU for higher intellectual powers: the first re- quii'es .accurate investigation of things and nice discrimination in classifying the subject; the second depends chiefly on a clear understanding and sti"ict attention to logical relations. The more minute the description and the more philo- sophical the dissertation, the greater will be the demand on the reflective and the reasoning powers. The double-ti-anslation, by placing in juxta- position the genius of two languages and of two nations, affords the professor a favorable oppor- tunity for enlightening his pupils on points of great interest, arising from the resemblances and differences which are rendered obvious by this twofold operation. The resemblances between two languages es- 186 THE STUDY OF LANG-IT AGES. tablisli the principles wMcli constitute general or comparative grammar; tlie differences between them mark ont the rules of the particular gram- mar of each. The comparison of several lan- guages, with a view to ascertain their lexico- graphic and grammatical affinities, throws light on their affiliation, on the migration of nations, and on the history of man. The professor, with- out touching on considerations above the reach of his pupils, will sometimes speak on these subjects, point out the origin of words and trace the modi- fications which they have undergone in passing from one language to another. If general grammar be j)roperly explained to young persons, at a time when the mind is capa- ble of such a study, there cannot be a doubt that it will open before them a large field on which they may exercise their reasoning faculties. Rising, therefore, with them above the facts, the generalization of which constitutes the art of grammar, the professor will easily enter on the consideration of the universal laws which govern languages and constitute tJie science of gram- mar. This high branch of literatm-e is too much neglected in scholastic instruction. Boys are ON MENTAI, CCLTTEE. 187 made to learn Latin, Greek, Frencli, and German grammars, but are seldom tanglit the laws whicli are common to aU languages, in contradistinction firom those which are peculiar to each. The foreigner, who is not competent to give this higher kind of instruction, and cannot im- prove Ms pupils in their own language, from not knowing it thoroughlv, should confine himself to his special sphere, and follow the processes which we have above described, in order to give them a practical knowledge of his own language, so as to make it for them an instrument of thought. If, however, he is versed in the grammar of his own language and in literary criticism, let him occasionallv take up one of the standai'd works which his class have read, and, after having ascertained that it is perfectly under- stood, let him make them analyze it as regai'ds the words, their nature, inflections, roots, pro- nunciation, derivation, synonymy, and different acceptations ; let him assist them in inferring tlie rules C'f grammar, from the recurrence of the same forms of expression. He will afterward, at a more advanced stage, tm-n theii- attention to style, point out the force and propriety of terms, the precision, elegance, and harmony of periods. 188 THE STUDY OF LANGUAGES. every tiling, in short, wMcli constitutes literary merit. Let the professor guard against the degener- ating of these considerations into mere parsing, an exercise which occupies so considerable a place in public instruction, and which, confined, as it usually is, to merely technical terms, aids in no way to the cultiyation of the mind, or to the practical acquaintance with a language. Particular grammar is an inductive art ; and, in all such arts, we arrive at principles from facts : the more numerous these are, the more general the rules. Custom is the law of language, gram- mar is only its generalization. Thus is grammar made, and thus it must be learned, from the lan- guage ; not the language from the grammar. All the rules of grammar are in the wi'itten page ; it is the teacher's office to bring them out, carefully avoiding abstract formulas, which chil- dren understand so imperfectly and forget so easily. If the latter had previously learned the rules, they would be deprived of the exercise in observation, comparison, analogy, and generali- zation, to which reasoning by induction leads. Moreover, rules which apply to tmknown facts, are pure abstractions devoid of interest ; -whereas, ON MENTAL CULTUEE. 189 tlie mind delights ia classifying scattered notions, and discovering tlie reason of known facts. This inductive or analytical mode of studying grammar, similar to the intellectual process by which we arrive at a knowledge of natural laws, is the most rational and the most favorable to naental discipline: it consists in observing facts, comparing them, remarking their resemblances and differences, and afterwai'd bringing into the same class all similar facts. Those which may be generalized constitute the rules, and those which are not comprised within any class form the ex- ceptions. The exercise of phrase-making, more especially ' by multiplying the expression of thought, aids in giving a practical knowledge of grammar. The construction of phrases after a model is the appli- cation of syntax to the expression of thought, for there is no phrase which does not exemplify some rule of syntax. It may be made by a judicious instructor the source of much grammatical in- formation to his pupils. In classifying and gen- eralizing constructions, formed on a principle of analogy, the laws which govern them are natu- rally evolved by induction. Practice and theory mutually aid each other, and grammar is thus 190 THE STUDY OF LANGUAGES. learned, not by an act of the memory, but by an act of judgment. Beneficial as these grammatical inductions may be to a learner, he should not be refused the instrument which may help him in his observations. It is not enough that he should infer the rules of composition from the phraseology : this incidental way of learning the grammatical principles of a language would never give him a complete and systematic knowledge of them. A good treatise on the subject is indispensable, if he wishes to have a comprehensive view of the theory. A few months assiduously devoted to this study would suffice to methodize and complete the scattered notions of grammar, acquired by induction ; and, if he aspires to being a grammarian, he should take up different treatises on the subject, read, compare, and judge for himself. A familiarity with the national grammar will be the best preparation for a similar study in the foreign language, as the learner will find in the grammar of that language the same technical denominations, and the same definitions. It renders more intelligible the explanations of the professor, who is often obliged, even at the outset, to advert to grammatical distinctions. ON MEXTAl CrLTrRE. 191 It also assists in translating from tlie native into the foreign tongne, because, in order to ascertain what is the foreign expression corre- sponding to the native, one mnst know the nature of the words to be translated and their functions in the sentence. On a cleai* concep- tion, for example, of the pei-son, tense, and mood of the English verb depends the correctness of the foreign one. If the learner has to render in French that part of an English verb which ends in ed, he must be able to distinguish, in every case, whether that verb is in the preterite or past participle, since it is differently translated, according as it is one or the other. Again, if hut be the word to be rendered into French, he must discriminate what part of speech it is in each particular instance ; that is, what is its pre- cise import; because, in its triple office of an adverb, a preposition, and a conjunction, it admits of different translations, viz., 7i« que, ex- ct'jife, mats. In addition to words belonging to different classes, there are a great many which admit of various equivalents in other languages, according as they are used in a literal or a figurative sense. A.3 a general rule, a word is iatei-preted in another 192 THE STUDY OF LANGtrAGES. language in as many ways as it admits of differ- ent significations. Althougli deficiencies and irregularities abound in all languages, they seldom occur on the same occasions in any two of them. No phrase is ren- dered literally, the constituent parts of which are, in their arrangement, relations, or meanings, in- consistent with the idea expressed, or with the laws of language, as, for example, the following idioms : He was offered a situation (familiarly used for a situation was offered to him). How do you like the book ? (the word how signifying in what manner, and the whole phrase implying that the person questioned does like the book, constitute a double inconsistency with the idea meant to be conveyed). I wish I was there now (a past tense used to mark the present). To wait on, to call upon, to hear from a person (three verbs used anomalously). From these observations, it is obvious that a thorough knowledge of the national idiom must jbe a great help toward acquiring a foreign living tongue. But classical education, which, as was keen, is intended not so much to teach a second language as to improve young people in their own, does not require any extensive knowledge ON MENTAL CULTUEE. 193 of the latter as a preparation: tMs education, on tlie contrary, secures it, under tte direction of an able instructor. With, his assistance all grammatical irregularities in the vernacular, all anomalies which might otherwise escape notice, are elicited by submitting them to the analytical process of translation : and the learner, being thus led ta inquire what are the different ideas attached to the same words, and what native expressions do or do not conform either to the idea intended or to the general principles of grammar, acquires a habit of nice discrimination and a critical knowl- edge of his own idiom. Under any circumstances, the usual order of study should be altered: the foreign grammar must be transferred from the lower to the higher classes, while the national should be taught as a foundation for linguistic studies ; the lower classes might continue to be denominated the grammar classes. Grammatical studies, although they do not necessarily impart the power of expression so effectually as the imitation of the great models, furnish the student with the means of entering into the secrets of composition, of exploring the mysterious laws of creative genius, and of sub- 194: THE STUDY OF LAKGTJAGES. mitting his own productions to the- control of reason and of establislied principles. It is then that theory becomes a useful auxiliary to practice.' When the learners haye completely mastered the art of hearing, the professor should occasion- ally address them in the foreign language on Tari- ous subjects of instruction. These subjects should be selected in reference to the studies in -which they are engaged at the time, and more particu- larly to the higher departments of that language ; he may treat of its genius and comparative merit ; investigate its origin, rise, and progress; unfold its importance as a vehicle of thought, or as a store of information ;' comment critically on its best works, and examine its literature, considered either absolutely or relatively to the national literature of his pupils. Should he feel diffident in extemporaneous delivery, he may read to them from the most eminent writers in the foreign lan- guage passages which would enrich their minds with useful knowledge and familiarize them with a pure and elegant style. In reading the Greek or Latin poets to his advanced pupils, the professor should avail him- self of the superiority of ancient prosody, to point out the effects of contrasted sounds, of long and ON MENTAL CULTtTKE. 195 short, Mgh and low notes; lie should explain to them the principles of quantity, accents, cadences, caesuras, rhythms, metres, pauses, all that consti- tutes the mechanism of verse and the melody of language. "We have seen that there are three modes of proceeding in learning languages: 1. The ex- clusively practical method, that of the mother tongue. 2. The exclusively comparative method, that which suits the study of Greek and Latin. 3. The practice-comparative, that which is re- quired for foreign living languages. The direct association of ideas with their, signs is the essence of the first. Translation from either language into the other is the essence of the second. By the third, the learner passes from comparison to practice, from the indirect to the direct use of the language. Although the comparative method can never completely secure the full mastery of a language, it wiU nevertheless remain the privilege of classi- cal instruction. It is the only one which permits an enlightened professor to give his pupils the greatest benefit of his own acquirements and his superior knowledge of their language. As for the practice-comparative method, while it shares 196 THE STUDY OF LANGUAGES. in the benefits of the other two, its special oflico is to make the foreign lanniuigo avaikible for the exchange of ideas : it adds little to man's intellec- tual power. Classical learning, as the most important in the education of youth, claims the lai'gor portion of the time apportioned to literary studies. With regard to modern languages, if their study is com- menced at a suitable age, eighteen months or two years at the utmost would suffice for gaining that jiractical acquaintance with one, whicli would enable a student to read it with pleasure and con- verse in it with ease. CHAPTEE VII. OliT EOUTINE. " Gardez-vous de la routine, c'est la mort de 1'eiiseignemen.t." Matter, " Beware of routine, it is fatal to teaching." A METHOD, in order to be rational and effi- cient, must be consistent witb the end proposed. We learn a foreign language for tbe sake of mat- . ing it tbe instrument of thought in the interna- tional exchange of ideas ; but, of the four arts by which this exchange is effected, the first two are indisputably of far higher importance and more general application than the other two, both as an end and as a means. The method ought then, first, to teach these two arts. Yet, strange to say, people will persist in saying that the principal object in studying a foreign living language is to learn to speak it; forgetting, moreover, that it must be understood before it is spoken, and that, to write well, one must have read much. This popular error has given birth to nearly 198 THE STUDY Olf 1-ANHUA(Uifl. all tlio luctliodH ill V(ii;'iio. l'\)i' i\w nioKt part tlioy aim at tho rajiid ac([iiiriMiuMit nf Unit art, and of that idoiio. Ootiti'iiiuinj;' liio (irdi'i- and wiso hIow- nowM of niduro, tlicy Itrunk (ill' liici olialii which (HHiuoctM toi;'(jlhoi- tho iuur f;'rcat ohjocts ol' a lan- gnaj;'in wiili loiiniin!;- nilort is a ^tohk error" (C'Oiidilliic). "I Avould fain liiuo any onu luuuo to mo lliiil'. (()ni;'iu', Avliic'li any ono vn\) learn, or speak an ho hIiohIiI do, l)y tlui rnlen of j^'rannnar" (Loeke). "An a|!,'e of llioory, ol' imre iJieory, Avonld nol: advance ti lierson one nlep in llie kiio\vledfj;e of n Inn^Mia^'e ; il. Avonld not tea(Oi to traiislaie a plirane " (IjO- niare). " Tlio rulert of j!;ra,iuina,r wineli arc re- sults, denioimtrated I'or liiiu who already ktiowH the hi,n<;'nax(da.iin9 Pluc.lic, " loiifj; remain it!;norant that there are in tho world Hiicli IhingH as gram- ma.i's I " (laii f;'ramniar ho consid(>red iw a saJe and in- ic^iligihle guide, when the graimiiariaiiH themselves are hut iiidifhaHMit sjieakers and writers? 'I'licir detinil-ions and rules arcrfor the most pa.rt ohsciire, iiicomiilete, a^iiil erroneous. Moreover, fho diver- sity of opinion whi(^li exists hetweeii them, n\: once shows the dilllcnlty of tho snhjiH^t and Iho ON EOtFTTNE. 203 absence of clear notions on the theory of gram- mar. In the whole circle of the arts and sciences there is not one in which the definitions and clas- sifications of their different authors are more dis- cordant and more tmcertain. The most element- arr questions have not jet receiyed. solutions whicli place them in the nimiber of universaJlj admitted truths. "Grammatici certant et sub judice lis est " (QuintiUian). Those who express themselves best in their own language owe their superiority far more to their own reflections than to the precepts of gram- marians. There was no methodical treatise on grammar at the time when Shakespeare, Milton, Dryden, Addison, Pope, Johnson, formed their style in writing. The same remark holds good with regard to Cicero, Virgil, Horace, to Moliere, Pascal, Comeille, La Fontaine, Boileau, Eacine, to Dante, Petrarch, Boccaccio, and many other celebrated writers, who, so far from having learned any thing from grammarians, supplied them with the materials from which they inferred their rules. One of the greatest mistakes arising from the lalse definition of grammar, a mistake against which the public cannot be too much on its guard, 204 THE STUDT OF LANGTTAGES. is that committed by certain persons wlio, know- ing only tlie grammar of a foreign language, set up as professors of the latter, under the pretence that teaching the grammar is teaching the lan- guage itself. Many compilers of grammars for beginners, put forth similar pretensions, as is ob- vious from their giving such books the title of " Method," for teaching, etc. (the language of which they treat). It frequently happens that grammar is not taught, but imposed as a lesson to be learned by rote, on young children incapable of understand- ing it. The verbatim repetition of the text is even sometimes insisted upon, so that, under this implicit injunction to attend to words rather than to sense, they seldom make an effort to compre- hend what they learn. " ]^othing," says Condil- lac, "is more useless than to weary a child by loading his memory with the rules of a language which he does not understand. Of what use is it for him to know rules by heart, if it is not in his power to apply them ? " ITo set of iTiles committed to memory will either form a profound scholar, or, what is in- finitely more important, create habits of patient observation and judgment. A man might be ON EOTJTINE. 205 acquainted with tlie results of many profound inquiries in all the various sciences; he might take them on credit, and act as if he believed them to be true ; but his understanding -wonld not be one jot advanced above that of an uninstructed workman. If the knowledge of all facts and the conclusions of aU researches could be poured into a man's mind without his own labor, he would really be less wise than he who has been properly trained to work the rule of simple pro- portion. It is not the letter, but the spirit of the rule, which can be productive of benefit. In grammar, as in morals and the sciences, it is impossible to apply a rule or to reason from a prraciple, unless we enter into the spirit of it. The most accurate rule, the wisest precept, if accepted without being understood in all its bearings, can never be ap- plied with perfect fitness in all possible circum- stances ; it will even become a contiaual source of errors. Let us, then, hope that we shall soon see ban- ished, from every school, " a method which," as , De G^rando observes, " is in direct opposition to the nature of things, which besets with abstrac- tions the novitiate of a mind as yet unprepared 206 THE STUDY OF LANGUAGES. for them, and which enters upon the study of a language, by the very notions which the knowl- edge of that language alone can give." All the objections to grammar for beginners apply equally to the written exercises, their usual auxiliaries. They cannot, any more than gram- mar, facilitate the apprenticeship to reading, for the understanding of a written text does not in any way imply the power of writing. Common sense requires that the learner should read before he writes, so as to know what is the best usage, in order to conform to it. It is, besides, contrary to reason to force chil- dren to compose in an idiom in which they will perhaps never have occasion to write, when they make comparatively so few efforts to acquire this same talent in their own, which would be so use- ful in the course of life, and so favorable to intel- lectual culture. The intimate relations between the thought and the style render coiuposition a highly intel- lectual exercise, only when the language is for the writer the direct and spontaneous expression of his ideas, and when he is practically conversant with its genius and phraseology, as is the case with the vernacular. But the learner who trans- ox EomxE. 207 Ir.tes into a tbrdgn idiom not ret fanuliar to him, does not -think in it, and is eren niiable to choose the words which wonld best conreT his ideas, be- cause he knows not their true import, nor the various shades of meaning which they conreT ; his consideration of words does not go beyond their orthography, their concord, or their respec- tive plac^ according as he is directed by the roles, which he has previously learned, or has before his eyes — a purely mechanical process, not much above a culinaiy operation from a cookery- book. These premature attempts at writing, as they do not pennit him to exercise his imitative or ini- agioative powers, and fraught with errors as they must be, are calculated to vitiate rather than im- prove bis taste. It is utterly impossible, as it has been erroneously believed, that they should culti- vate his understanding, or impart to him the power of discovering and appreciating t]ie beau- ties of foreign literary productions. They seem, like most other contrivances of routine, to have been introduced, in order to aSord the instructor an opportunity of correcting errors which could be avoided by processes more conformable to reason. 208 THE STUDY OF LAKG0AGES. They are condemned by all all writers who have treated of linguistic studies. Eollin, timid as he wag in his educational reforms, says : " To write Latin well, one must know the turns, the idioms, the rules of the language, and have at command a considerable provision of words, the force of which is felt, and the just appli- cation of which can be made. Now, all this can be done only by explaining authors, who are like a living dictionary and a speaking grammar, in which are learned by experience the force and the true use of words and phrases, as well as the rules of syntax. I have no hesitation in declaring that, in the beginning, written exercises in the foreign tongue ought to be entirely excluded, as being calculated only to torment young people by painful and useless labor, and to inspire them with dislike to a study which generally di'aws on them nothing but reprimands and punishments." Unwilling to swell this volume with quota- tions, we will be content with naming, among those who condemn the practice, Hoger Ascham, Milton, Locke, the distinguished scholars of Port Eoyal, Montaigne, Dumarsais, Pluche, Eadonvil- liers. La Chdlotais, Diderot, D'Alembert, J. J". Eousseau, Bernardin de St. Pierre, Guizot. OS KOTTCfB. 209 Soiue p^otisans of itratine, unxdUiug to reject the method of §^»iunatic»l eserei^ althongli awsie of its d^Efe<^ haT© compiled luaiinak, in which they conteiTe eT«y po^hle means: toiSiciK- tat« the oibservance of the itil^ aad spare Toung people Ao troTible of reflectai^: ther gire not oaly tiiie ft«a^ words, bnt their order ; ttter in- dicate the gender and nxnnheff, mood and tense ; thoT point out whiai wointe 4^ *<* ^ omittied or $^p£ed. Leaineats mechaoicalhr avail them- s^lviK of ttife afstetance without inqnirii^ into €he di^sfsico of the idiom hetwe«x the two hm- gnageei, olt^i exeax without reading the rule be- jRsre thoT write the esCT^e, and without attaid- ii^ to fl»e idea to be esjarefised ; s\> that ti»e resn- edy ij wvwse titan the exil. XexTts- have these uncoimected eoaupositioiis h;d to the fissnation of a sood s»\ie. Inappropriate and inefficteat a$ ax^f grammati- cal e:xa^3e$ £.^ aeqnirii^ the art of compo^tion in a fot^pft language, thev are, in ateuiditT, iar s«ipa£sied br the ptaedce of tiaielating into it ftom a nat&aaal author at an eari v period of the stjsdv, wheffl the kamer hse no rules, no model to snide him; whaihe knows neithes- the diififar«it a«?«f**tioi:e of the foreign woivfe nor Ae shadt^ 210 THE STUDY OF LANGUAGES. of difference between synonymes; wlaen he cannot even imagine whether that language admits, or does not admit, of rendering literally the forms of another ; whether it has or has not equivalent idiomatic or figurative expressions. It is so pre- posterous, that it defies all argumentation. Even at an advanced period, it is a most injudicious practice. To translate into the native idiom, pre- sents, as has been aten, great obstacles to those most conversant with it; how, in the name of common sense, can it be done in a language with which one is not acquainted ? It is an absurdity of the same kind as that of reading aloud before having heard the words. The cause of this anomaly may be found in the fact that foreigners who teach their own lan- guage, being at first little versed Jn that of their pupils, feel their inability to translate into it, and resort to the opposite practice. Their correction of compositions in their own language gives an appearance of usefulness to their services ; but, in reality, they only delay the progress of their pupils, in the first three arts, without teaching them the fourth. Whether from pride or weak- ness of mind, men seldom think of leaving off a false course on which they have entered. ON EOTrriNE. 211 They fall into error throngh ignorance ; they re- main in it from habit. The time which the unfortunate victims of routine spend at this work* in the absence of the instructor, leaves them but little leisure for read- ing ; and, on the other hand, the correcting of the tasks in class takes up a considerable portion of the master's time, which would be much better employed in explaining to them the ancient clas- sics, or in enabling them to understand oral ex- pression in living languages. They read, in the intervals of their lessons, only as much as their instructor has time to hear them translate in class, which is far from being sufficient for acquiring the art of reading or becoming acquainted with all the forms of the written language, and especially with its orthography. Learners, not obtaining from example any as- sistance toward acquiring this elementary depart- ment of composition, are made to resort to special exercises, more or less irrational, as they reject the commands of nature. Among others, we may mention dictation, a practice altogether inefficient. In Italian and Spanish, for instance, the conformi- ty of the spelling with the pronunciation renders it utterly useless. So uniform is the representa- 10 212 THE STtJDT OF LANGUAGES. tive power of ; the letters in these languages, that to pronounce an Italian or a Spanish word is to spell it. In German, dictation is not much more useful, because the same letters representing the same sounds and articulations, it suffices to know the power of their alphabetical characters, in order to deduce the spelling of the words from their sounds. French orthography, which is in frequent disagreement with its pronunciation, should be learned by a reference to its orthoepy, etymology, and syntax, rather than by dictation. As for the English language, which is rarely spelt as it is pronounced, its orthogi-aphy is the fruit of practice alone. He who knows the speUing of an English word derives no benefit from writing it, and he who is not previously acquainted with it, will seldom be able to spell it from hearing. Dictation, like reading aloud, is a test of pro- ficiency, not a means of teaching. Although, however, it cannot prevent the commission of errors, it afibrds the means of detecting and cor- recting, them. But, viewed even in this light, dictation should be resorted to cautiously; be- cause, for one word that the pupil may thus learn to spell, he wastes time in writing many ON EOTTrmE. 213 » ■which he knew before. This is purchasing too dearly a species of information which can be easily gained conjointly with higher departments of composition. For those who make reading the basis of study, and follow our suggestions, correct spelling is as necessary a consequence as is pronunciation in the acquisition of the mother- tongue. By dictation, the professor teaches noth- ing to his pupils that they could not learn by themselves. It is so universally adopted, only because it is easily available, demanding, on the part of the teacher, little trouble, capacity, or in- formation. Premature exercises of pronunciation, forced in by the priority given to the art of speaking, have a still more pernicious effect than those which relate to orthography. They are primarily addressed to the ei/e, as they proceed from the letters to the sounds; whereas, on the contrary, a knowledge of the pronunciation should first be gained, in order to ascertain the value of the al- phabetical characters which represent them. It is the inversion of the order in which the different branches of a language should be learned that has given rise to the practice of reading aloud, and to all those dissertations on- the letters of the 214 THE STUDY OF LAITGITAGE8. alphabet wMch serye as an introduction to nearly all grammars. As an early initiation in oral reading, books have been contrived, in which the prontmciation of the foreign language is assimilated with that of the native, by alphabetical combinations in the latter. The attempt to spell words in one language as they are pronounced in another, must, in most cases, prove unsuccessful; for the pen can neither represent new sounds to the eye, nor mark the imperceptible shades of colloquial intonation. Every language has vowel sounds, articulations, and an accentuation peculiar to it- self. Of the French vocal elements, for example, seven sounds and one articulation, are not in the English pronunciation, and cannot, therefore, be represented by English letters.* Such contriv- ances only famiharize the eye with a defective spelling of the foreign words. There is nothing more common than to make beginners read aloud each phrase of the text before translating it. In Latin, which is pro- nounced by modern nations nearly the same as * The French sounds represented by S, u, eu, an, in, on, un, and the articulation of which ill is the sign, do not exist in the English pronunciation. ON EOXITINB. 215 their own language, the previous reading has no inconvenience; it is even necessary, in order to unravel the inverted arrangement of its words; and to construe these conformably to the genius of the language into which the translation is to be made. The case is different as regards the mod- ern languages of Europe, in which the words often assume the same order, but differ widely in their pronunciation. "Without in any way helping to understand the text, this mode of proceeding inevitably leads to a defective pro- nunciation. Alternately pronouncing and translating each sentence, constantly disjoins the subject, and thereby not only lessens the interest that the nar- rative might create, but also throws an obstacle in the way of making out the sense from the con- text. Besides, a beginner cannot attend at the same time to the pronunciation and the construc- tion, both being new to him; he necessarily neglects the one while attending to the other. Finally, this practice forms habits contrary to the object most desirable in translating, the power of doing so at sight and without prepara- tion. In a class, reading aloud engages only one 216 THE STUDY OF LANGUAGES. person at a time, and leaves all the others for a great part of the time in a state of lamentable listlessness, or, which is stiU worse, if they listen, accustoms them to the more or less defective pro- nunciation of those who read. Eeciting and correcting eJcercises in class are liable to a similar objection : they require the exclusive attention of the master for each pupil separately. When reading aloud is practised at the com- mencement of the study, in the presence of an instructor, careful to correct every error of pro- nunciation, these recur so frequently that very little time remains for translation, which, at the outset, ought to be the only object of considera- tion. These difficulties would be only partially obviated if the professor himself, as is sometimes done, should pronounce every word or phrase before the pupil; for the ear cannot, on a first hearing, notice at once all the shades of difference which mark the vocal elements in a strange lan- guage, and especially the accent and quantity; these are so delicate that, to be perceived, they demand extreme sensibility of the organ, culti- vated by long and patient practice in hearing. "With regard to the art of reading as an ac- complishment, its acquisition is utterly impossible ON ROUTINE. 217 in a foreign language, so long as a person has not become so habituated to its pronunciation, as to have his mind completely free to give his at- tention exclusively to the author's thought. It is, moreover, of little service ; for a person has rarely occasion to read in a foreign language to his own countrymen, and still less to foreigners. But, in one's own language, no pains should be spared in acquiring this accomplishment, which, though most useful and agreeable, is seldom pos- The adoption of these different processes is injurious to the teaching of languages, especially because it favors the ignorance and incapacity of those who resort to teaching, after having failed in other occupations. The exercises of memory, in which the master acts a merely passive part, are equally objectionable. Teachers are apt to resort exclusively to this faculty for various branches of instruction, although there is no need of any special exercise to call it into activity ; for its action, like that of attention, is comprised in that of the other faculties. It is by no means unusual to see unfortunate children poring over vocabularies or other compilations, forced to learn by heart lists of words, conjugations, dialogues. 218 THE STDDT OF LAITGUAGES. passages of books, wMcli they never turn to ac- count for the expression of their thoughts. The study of words as an introduction to a language, is a violation of the laws of nature, and leads to no useful result. One could read two or three volumes with interest, by means of interpretations in juxtaposition, in less time than would be consumed in painfully learning five or six hundred words as a preparation for reading. But this point has already been alluded to ; we will now advert to dialogues and extracts as mne- monic lessons. In these the attention is directed solely to the words which are associated in the mind by their contiguity. By dint of repetition they are necessarily recalled in their order of suc- cession, each word suggesting that which follows. The more frequently the lesson is repeated in the act of learning it, the easier is the recitation, the more also does the text escape analysis and the control of the will. Recitation is an exercise in oratorical delivery, not in the expression of one's own ideas. The habit produced by - the repetition of a text is diametrically opposed to the mental opera- tion required for the expression of thought: speaking is an act of the judgment, reciting an ON EOITTINE. 219 act of the memory. The first consists in associa- ting words with the ideas as these arise in the mind; the second in merely associating words with each other on the principle of contiguity. Incessant change of words and phraseology char- acterizes the one ; immutability of form and order is the essence of the other. "We command the former, we are slaves to the latter. In speaking, the attention is intent on ideas ; in reciting, it is intent on words ; in the former the words are sub- ordinate to the ideas, in the latter the ideas are subordinate to the words, very often they are not taken into consideration at all ; there is nothing so common with children as to repeat what they do not in the least understand. " To know by heart is not knowing," said Montaigne. Dialogues, like extracts learned by rote, teach to recite, not to converse. "Whatever be the num- ber with which a learner has loaded his memory, he is only the tame repeater of another man's ideas ; he is never called upon to express his own, and his power of conversing is regulated by the whim and peculiar notions of the compiler. It is obvious that the art of speaking depends not so much on the recollection of a large number of dialogues as on the power of spontaneously 220 THE STUDY OF LANGUAGES. constructing sentences suited to the ever-changing circumstances of social intercourse. Analogy, the power through the instrumentality of which command of expression is acquired, is therefore more effective than mere recollection of phrase- ology. How can it be supposed that a dialogue, for example, between a lady and her dress-maker, written most likely by a man little conversant with the caprices of fashion, or with female attire, could serve as a type for all conversations between ladies and their dress-makers, despite the changes of fashion and whatever be the season of the year, the dispositions, ages, wants, taste, wealth of the parties, and innumerable other circum- stances. The victims of this system, however, are generally spared the trouble of testing the usefulness of these dialogues ; they generally forget them long before they have an opportunity to turn them to account. Not only are these lessons of very little ser- vice, but, as they demand much time, and are very irksome, they can hardly fail to inspire aver- sion for study. They do not even cultivate the memory in any useful way, for the faculty of remembering words in a given order, serves no ON EOUTINE. 221 practical pm-pose, unless to actors in learning their parts. 1^0 faculty can be exercised and improved generally by any particular process ; its cultiva- tion in one direction does not extend its power in another ; thus persons, practised in musical modu- lations, have no superiority over others in catch- ing the pronunciation and accent of foreign lan- guages. If the sight be exercised on colors, it will not better appreciate forms or distances, and if exercised on either forms or distances, only a similar partial improvement will be produced. It is the same with the intellectual faculties. Their development is always in accordance with the means by which it is attained. The person who has been much engaged in learning mere words or passages of books . will not, from that special exercise, possess greater power in recollecting facts, localities, dates, the subject-matter of a book or of a speech. In short, the practice of committing to memory words and phrases, gives to this faculty nothing more than an aptitude for pan-oting another man's words and phrases, and such an aptitude will never raise its possessor in the scale of intellect, or enable him to carry on more successfully the affairs of life. The 222 THE STUDY OF LAJStgtjAGES. meclianical memory of words should not be made a substitute for the intellectual memory of tbings. All tbe time which a boy spends in learning by heart and reciting lessons is lost for the exer- cise of his judgment and for the practice of the language. In classes the great majority of its members remain unoccupied, awaiting their turn to be heard. As for the professor, what does he do ? He Hstens, he does not teach. However ex- tensive his knowledge may be, it is a dead letter to his pupils. He who, in his teaching, does not go beyond the books, is no professor. The numerous tasks which, in the prevailing systems of instruction, young people often -have to prepare for their instructor, allow them no time to practise reading in his absence, or the spoken language in his presence. ' They leave school, for the most part, without having entered upon any essential department of the study. They are given to understand that their labor is over when they begin to read aloud, and translate or parse the foreign language with fluency, when they have conjugated all the verbs, and repeated a volume of dialogues, when they have learned all the rules, and written all the exercises of their ON EOUTINE. 223 grammar ; and yet nothing of all this is the lan- guage, the practical language. These sad results are the more remarkable as we every day see young children acquire a second language abroad with a facihty which would put to shame an adult who was learning the same language in the plenitude of his reason, who, having only to collect signs to be attached to the ideas he possesses, spends considerably more time in learning them than do these children in learn- ing to speak the language, although the latter have to acquire every thing — the ideas as well as their signs. Nothing shows more obviously the fallacy of the methods pursued. The easiest and most useful branch, the art of reading, they seldom acquire so as to turn it to advantage in after-life. Their study of it does not extend beyond the translation of a few vol- umes or fragments of volumes ; and, consequently, there are very few who, after leaving school, ever open a book in a living language, unable as they are to read that language with ease or pleasure. These are the sad finiits of a routine which, by confining learners exclusively to translation, incapacitates them from entering into the spirit 224 THE STUDY OF LANGUAGES. of authors, or to appreciate their merit, and wMcli, by frequently enforcing on tliem the practice of scrap-reading, leaves them lamentably ignorant of the foreign literature. The routine of classical studies especially is deplorable in its consequences; it does not con- form either to the laws of ligature or to the re- quirements of society, sets at naught spontaneity, curiosity, imitation, and analogy ; overloads mem- ory to the prejudice of judgment, reverses the order prescribed by reason, ia passing from the signs to the things signified, from the rules of grammar to the facts of language, from the art of writing to the art of reading ; nor does it fally secure the intellectual development aimed at by these studies, and keeps young people too long engaged on things of the past, to the exclusion of the knowledge which the progress of civilization has rendered indispensable. We do not mean by these remarks to cast censure on the profession of teacher. The most enlightened among its members lament the pres- ent state of things as much as we do, but are unable to offer a remedy. So strong is the preju- dice in favor of the course sanctioned by time and school-routine, that the head of a private school. ON EourmE. 225 or the professor who should wish to introduce a reform in this respect, would probably be exposed to great personal prejudice. " Custom," says Eollin, " often exercises over minds a sort of tyranny which keeps them in bondage and hinders the use of reason, which, in these matters, is a surer guide than example, how- ever authorized by time." Of all the known methods there is not one in strict conformity with the process of nature or which furnishes the means of thoroughly acquir- ing a foreign language, in its fourfold applica- tion. Dumarsais, Eollin, Pluche, Kadonvilliers, and Lemare, authors of special treatises on the art of teaching languages, have presented very judicious suggestions on the matter; but, as they had ex- clusively in view the study of the dead languages, they suggested no means of effecting the oral exchange of ideas. The method of interlinear translations, advocated by some of them, as well as by Locke, however excellent and practical, can lead only to the understanding of the written lan- guage. The imitation of a model-book, as recommend- ed by Jacotot, certainly places the student in the 226 THE STUDY OF LAI^GUAGES. right track ; but it is a perversion of his princi- ple, "All is in all," to limit, as he does, the knowledge of a language to the study of a single volume, and to force on his pupils a defective pronunciation, by insisting, as he does, on their committing this volume to memory at the very outset, when they are utterly ignorant of the pronunciation of the foreign tongue. This is not surely the way in which !N^ature proceeds. He initiates his pupils in the art of speaking by ask- ing them questions on the text of this volume, which they are to answer by repeating the very words of the author. This repetition of words and phrases, drawn exclusively from one text, is assuredly insufficient. The subject-matter and the phraseology of his model-book, Telemachus, are also ill adapted for the purposes of ordinary conversation. In the Ollendorff system, practice justly comes before theory; but the trivial phraseology of which its lessons are made up, and their recita- tion, afford no aid toward extempore speaking or the formation of a good style, despite the preten- tious title of his book : " Method of learning to speak, read, and wi'ite German in six months." This title alone proves that Ollendorff did not ON KOTTTINE. 227 even suspect that tlie art of understanding the spoken language could be taught. Eohertson's method, partly copied from Jaco- tot's, is not less barren in results. Eejecting the progressive order prescribed by Nature, it aims, from the outset, at the fom- arts simultaneously; hence his premature lessons of pronunciation, grammar, etymology, parsing, analytical and philological disquisitions, which constantly take the attention away from what is practically use- ful. The pupils have no other subject of study, in the whole course of this method, but a text still more restricted than that of Jacotot; and, far from recommending the reading of good au- thors, Mr. Kobertson pretends, contrary to com- mon sense, that the short Persian tale of three or four pages, on which his lessons turn, suffices for aU the acquirements of the foreign language. In short, neither these methods, nor any other that has come to our knowledge, secure the means of understanding foreigners when speaking their own language, and still less that of acquiring the power of thinking in that idiom, a power essential to its thorough mastery. The diversity of lights in which linguistic in- struction is viewed, owing to the absence of a 228 THE STUDY OF LANGUAGES. rational classification, has naturally produced a corresponding diversity in the mode of effecting it. Every teacher, in entering on his profession, bewildered by all these processes, many of which are in direct opposition to one another, has to contrive a method for himself, or he must blindly follow the routine which often has no other recom- mendation than its antiquity. This appeal to our fathers in what regards education keeps the mind in bondage, and is an obstacle to progress : they are our juniors in the world. "We have our own experience in addition to theirs, and start in life with greater advantages" ; we consequently ought to know more, and cannot make their notions or opinions the standard of our conduct. It is time to reject the worn-out ma- chinery of our forefathers. Let us apply to mind, as we have done to matter, new powers and new processes. Let a rational system of learning lan- guages bring men of all nations into communion, as steam has brought them into contact.* * We refer the reader to the work mentioned in the preface for more ample information on the system here briefly sketched, and on the processes which constitute its practical application. THE END. LIST OF WORKS PUBLISHED BY D. APPLETOI( & CO., 90, 92 & 94 GBAITD ST., NEW TORK, A Descriptive Catalogxce, with full titles and prices^ trho/y he had, f on i About's Eoman Question. Adams* Boys at Home. Edgar Clifton. Addison's Spectator. 6 vols. Adler's German and Englisli Diction- ary. Abridged do. do. do. German Eeader. " Literatnre. Ollendorff for Jjearnlng German. Key to the Exercises. Iphigenia in Tauris. After Icebergs with a Painter. Agnel's Book of Chess. Agnilar"'s Home Influence. Mother's Eecompense. Days of Bruce. 2 vols. Home Scenes. "Woman's JYiendship. Women of IsraeL 2 Tols. Vale of Cedars. Ahn's French Method. Spanish Grammar. A Key to same. German Method- 1 voL Or, separately — ^First Course. 1 TOL Second " 1 TOL Aids to Faith. A series of Essays, by Various Writers, Aikin'a British Poets. From Chaucer to the Present Time. 3 vols. Album for Postage Stamps. Albums of Foreign Galleries; in 7 folios. Alden's Elements of Intellectual Phi- losophy. Alison's IkCseellaneous Essays. Allen's Mechanics of Nature. Alsop's Charms of Fancy. Amelia's Poems. American Poets (Gems horn the). American Eloquence. A Collection of Speeches and Addresses. 2 vols. American System of Education : 1. Hand-Book of Anglo-Saxon Eoot-W()rds. 2. Iland-Book of Anglo-Saxon Derivatives. American System of Education : 3. Hand-Book of Engrafted Words. Anderson's Mercantile Correspond- ence. Andrews' JTew French Instructor. ■ A Key to the above. Annals of San Francisco. Antisell on Coal Oils. Anthon's I^w Student. Appletons' New American Oyclopffidia of XJBeM Knowledge. 16 vols. Annual Cyclopffidia, and Eegister of Important Events for 1861, '62, '63, '64, '65. Cyclopsedia of Biography, For- eign and American. Cyclopedia of Drawing. The same in parts : Topographical Drawing. Perspective and Geometrical Drawing. Shading and Shadows. Drawing Instruments and their Uses. Architectmal Drawing and De- sign. Mechanical Drawing and Design. Dictionary of Mechanics and En- giueering. 2 large vols. Eailway Guide. American Illustrated Guide Book. 1 vol. Do. do., separately : 1. Eastern and Middle States, and British Provinces. 1 vol. 2. Southern and Western States, and the Territories. 1 vol. Companion Hand-Book of Travel. Arabian Nights' Entertainments. , Arnold's (S. G.) History of the State of Ehode Island. 2 vols. Arnold's (Dr.) History of Eome. Modern History. Arnold's Classical Series : First Latin Book. First and Second Latin Book ant^ Grammar. Latin Prose Composition. Cornelius Nepos. First Greek Book. Greek Prose Composition Book, 1. D. APPLETON & CO.'S PUBLICATIONS. Arnold's Greek Prose Composition Book, 2. Greek lieading Book. Arth ur's (T. 8.) Tired of Housekeeping. Arthur's (W.) Successful Merchant. At Anchor ; or, A Story of our Civil "War. Atlantic Library, 1 vols, in ease. Attach,6 in Madrid. Aunt iTanny's Story Book. ■ Mitten Series. 6 vols, in case. Night Cap Series. 6 vols, in case. Badois' EnglisTi Grammar for French- men. A Key to the above. Baine''s Manual of Composition and Ehetoric. Bakewell's Great Facts Baldwin's Flush Times. Party Leaders. Balmanno's Pen and Pencil. Bank Law of the United States. Barrett's Beauty for Ashes. Bartlett's U. S. Explorations. 2 vols. Cheap edition. 2 vols, in 1. Harwell's Good in Every Thiug. Bassnett's Theory of Storms. Baxley's "West Coast of America and Hawaiian Islands. Beach's Pelayo. An Epic. Beall (John T.), Trial of. Beauties of Sacred Literature. Beauties of Sacred Poetry. Beaumont and Fletcher's "Works. 2 vols. Belem's Spanish Phrase Book. Bello's Spanish Grammar (in Spanish). Benedict's Run Through Europe. Benton on the Dred Scott Case. Thirty Tears' View. 2 vols. ■ Debates of Congress. 16 vols. Bertha Percy. By Margaret Field. 12mo. Bertram's Harvest of the Sea. Eco- nomic and Natural History of Fishes. Bessie and Jessie's Second Book, Beza's Novum Testamentum. Bibles in all styles of bindings and various prices. Bible Stories, in Bible Language." Black's General Atlas of the w orld. Bloomfield's Farmer's Boy. Blot's "What to Eat, and Hotv to Cook it. Blue and Gold Poets. 6 vols, in case. Boise's Greek Exercises. . First Three Books of Xenophon's Anabasis. Bojcsen's Greek and Roman Antiqui- ties. Book of Common Prayer. Yarioiis prices. Boone^e Life and Adventures, Bourne's Catechism of the Steam En- gine. Hand-Book of the Steam Engine. Treatise on the Steam Engine. Boy's Book of Modem Travel - Own Toy Maker -* Bradford's Peter the Great. Bradley's (Mary E.) Louglass Farm^ Bradley's (Chas.) Sermons. Brady's Christmas Dream. Breakfast, Dinner, and Tea. British Poets. From Chaucer to the Present Time. 8 large vols. British Poets. Cabinet Edition. 15 vols. Brooks' Ballads and Translations. Bro'wn, Jones, and Robinson's Tour. Bryan's English Grammar for Gei'- xnans. Bryant & Stratton's Commercial Law. Bryant's Poems, Illustrated. Poems. 2 vols. Thirty Poems. Poems. Blue and Gold, Letters from S^ain. Buchanan's Administration. Buckle's Civilization in England. 2 vols. Bunyan's Divine Emblems. Burdett's Chances and Changes. Never Too Late. Burgess' Photograph Manual, 12mo. Burnett (James E.) on the Ihirty- nine Articles. Burnett (Peter H.), The Path which led a Protestant Lawyer to the Catholic Church. Burnouf B Gramatica Latina. Bums' (Jabez) Cyclopiedia of Sermons, Burns' (Robert) Poems. Burton's CyclopEedia of "Wit and Hu- mor. 2 vols. Butler's Martin Yan Buren. Butler's (F.) Spanish Teacher. Butler's (S.) Hudibras. Butler's (T. B.) Guide to the Weather. Butler's ("Wm. Allen) Two Millions. Byron Gallery. The Gallery of By- ron Beauties. Poetical "Works. Life and Letters. "Works. Illustrated. Cceleb's Laws and Practice of "Whist. Csesar's Commentaries. Caird's Prairie Farming. Calhoun's Works and Speeches. 6 vols. Campbell's (Thos.) Gertrude of "Wy- oming. Poems. Campbell (Judge) on Shakespeare. Canot, Life of Captain. Carlyle's (Thomas) Essays. Carreno's Manual of Politenesa, D. APPLETON & CO.'S PUBLICATIONS. Carreno's Gompendio del Manual de Urbanidad. Casseday's Poetic Lacon. Cavendish's Lawa of "Wliist. Cei'vantes' Don Quixote, in Spanish. Don Quixote, in English. Cesar KHistoire de Jules, par S. M. I. Napoleon III. Vol. I., "with Maps and Portrait. (French.) Cheap Edition, witiiout Maps and Portrait. Maps and Portrait, for cheap edi- tion, in envelopes. Champlin's English Grammar. Greek Grammar. Chase on the Constitution and Canons. Chaucer's Poems. Chevalier on Gold. Children's Holidays. Child's First History. Choquet's French. Composition. French Conversation. Cicero de Olficiis. Chittenden's Eeport of the Peace Con- vention. Select Orations. Clarke's (D. S.) Scripture Promises. Clarke's (Mi'S. Cowden) Iron Cousin. Clark's (H. J.) Mind in Nature. Cleavelahd and Backus' Villas and Cot- tages. Cleveland's (H. W. S.) Hints to Eifle- men. Cloud C^stals. A Snow Flake Alhum. Cobb's (J. B.) Miscellanies. Coe's Spanish Drawing Cards. 10 parts. Coe's Drawing Cards. 10 parts. Colenso on the Pentateuch. 2 vols. On the Eomans. Coleridge's Poems. Collins' Amoor. Collins' (T. "W.) Humanics. Collet's Dramatic French Eeader. Comings' Physiology. Companion to Physiology. Comment on Parle a Paris. Congreve's Comedy. Continental Librtu-y. 6 vols, in case. Cooke's Life of Stonewall Jackson. Cookery, by an American Lady. Cooley's Cyclopaedia of Eeceipts. Cooper's Mount Vernon. Copley's Early Friendship. . Poplar Grove. Cornell's First Steps in Geography. Primary Geography. Intermediate Geography. Grammar School Geography. High School Geography and Atlas. High School Geography. " " Atlas. Map Drawing. 12 maps in case. Outline Maps, with Key. 13 maps in portfolio. Or, the Key, separately. Cornwall on Music. Correlation and Conservation of Forces. Cortez' Life and Adventures. Cotter on the Mass and Rubrics. Cottin's Elizabeth ; or, the Exiles of Siberia. Cousin Alice's Juveniles. Cousin Carrie's Sun Rays. Keep a Good Heart. Cousin's Modern Philosophy. 2 vols, On the True and Beautiful. -^— Only Romance. Coutan's French Poetry. Covell's English Grammar. Cowles' Excaange Tables. Cowper's Homer's Iliad. Poems. Cox's Eight Years in Congress, from 1857 to 1865. Coxe's Christian Ballads. Creasy on the English Constitution. Crisis (The). Crosby s (A.) Geometry. Crosby's (H.) CEdipus Tyrannus. Crosby's (W. IT.) Quintus Curtius Eufus. Crowe's Linny Lockwood. Curry's Volunteer Book. , Gust's Invalid's Book. Cyclopedia of Commercial and Busi- ness /iUecdotes. 2 vols* D'Abrantes' Memoires of Napoleon, P- vols. Dairyman's (The) Daughter. Dana's Household Poetry. Darwin's Origin of Species. Dante's Poems, Dasent's Pop'-'^ t Tales from the Norse. Davenport's Christian ITnity and its Recovery. Dawson's Archaia. De Belem's Spanish Phrase-Book. De Fivas' Elementary French Reader. Classic French Eeader. De Foe's Robinson Crusoe. De Girardin's Marguerite. Stories of an Old Maid. De Hart on Courts Martial. De L'Ardeche's History of Napoleon. De Peyrac's Comment on Parle. De Staei's Corinne, ou L'ltalie. De Veitelle's Mercantile Dictionary. De Vere's Spanish Grammar. Dew's Historical Digest. Dickens's (Charles) Works. Original Illustrations. 24 vols. Dies Irae and Stabat Mater, bound to- f;ether. rae, alone, and Stabat Mater, alone. Dix's (John A.) Winter in Madeira. Speeches and Addresses. 2 vols, D. APPLETON & CO.'S PUBLICATIONS. Dix's (Rev. M.) Lost Unity of the Christian World. Dr. Oldham at Groystones, and his Tallt there. Doano's Works. 4 vols. Downing's liural Architecture. Dryden's Poems. Dunlap's Spirit History of Man. DuBseldorf Gallery, Gems ft-om the. Dvvight on the Study of Art. Ebony Idol (The). Ede's Management of Steel. Edith Vaughan's Victory. Egloffstein's Geology and Physical Geography of Mexico. i Eichhorn'fl German Grammar. ■ Elliot's Fine Work on Birds. T parts, or in 1 vol. Ellsworth's Text-Book of Penman- ship. Ely's Journal. Enfleld'B Indian Corn ; its Value, Cul- ture, and Uses. Estvan's War Pictures. Evans' History of the Shakers. Evelyn's Life of Mrs. Godolphin. Kvei'ett's Mount Vernon Papers. Fables, Original and Selected. Farrar's History of Free Thought Faustus. Fay's Poems. F6n61on's Telemaque. The same, in 2 vols. Tclemachus. Field's Bertha Percy. Field's (MO City Ai-chitecturo. Figuier's world before the Deluge. Fireside Libraiy. 8 vols, in case. First Thoughts. Fiji and the Fijians. Flint's Pljysiology of Man. Florian's William Tell. Flower Pictures. Fontana's Italian Grammar. Foote's Africa and the American Flog. Foresti's Italian Extracts. Four Gospels (The). Franklin's Man's Cry and God's Gra- cious Answer. Frieze's Tenth and Twelfth Books of Quintilian. FiiUerton's (Lady G.) Too Strange Not to be True. funny Story Book. Garland's Life of Eandolph. Gaskell's Life of Bront6. 2 vols. The same, cheaper' edition, in 1 vol. George Ready. Gerard's French Readings. Gertrude's Philip Randolph. Gesenius' Hebrew Grammar. Ghostly Colloquies. Gibbes' Documentary History. vols. Gibbons' Banks of New York. (Tilllllan's Literary Portraits. Gillespie on Land Surveying. Girardin on Dramatic Literature. Goadby's Text-Book of Phvsiology. Goethe's Iphigcnia in Tam'is. Goldsmith^s Essays. Vicar of Wakefield. Gossc's l^venings with the Microscope. Goulburn's Cilice of the Holy Com- munion. Idle Word. Manual of Confirmation. Sermons. Study of the Holy Scriptnrcs. Thoughts on Personal Religion. Gould's (E. 8.) Comedy. Gould's (W. M.) Zephyrs. Graham's English Synonymes. Grandmamma Easy s Toy Books. Grandmother's Library. C vols, in case. Grand's Spanish Arithmetic. Grant's Report on the Armies of the United States 18(i4-'66. Grauet's Portuguese Grammar. Grayson's Theory bf Chiistianity. Greek Testament. Greene's (F. II.) Primary Botany. Class-Book of Botany. Greene's (G. W.) Companion to OUen- dorfl'. First Lessons in French. First Lessons in Italian. Middle Ages. Gregory's Mathematics. Griffin on the Gospel. Griifith's Poems. Griswold's Repuhlican Court. Sacred Poets. Guizot's (Madame) Tales. Guizot's (M.) Civilization in Europe. 4 vols. School edition. 1 vol. New Edition, on tinted paper. 4 vols. Gurowski's America and Europe. Russia 03 it is. Hadley's Greek Grammar. Hahn's Greek Testament Hall's (B. H.) Eastern Vermont Hall's (C. H.) Notes on the Gospels. 2 vols. Hall's (E. II.) Guide to the Great West Ilalleck's Poems. Poems. Pocket size, blue and jold. onng America. Halleok's (H. WO Military Science. Hamilton's (Sir Wm.) Philosophv. Hamilton's (A.) Writings. 6 vols. go Yov D. APPLETON & CO.'S PUBLICATIONS. Hand-Boo^s on Education. Hand-Book of Anglo-Saxon Eoot- "Words. Hand-Book of Anglo-Saxon Beri- vatives. Hand-Book of the Engrafted Words. Handy-Book ot Property Law. Happy Child's Library. 18 vols, in case. Harkness' First Greek Book. Latin Grammar. First Latin Book. ■ Second " Latin Reader. Hase's History of the Church. Haskell's Housekeeper's Encyclo- predia. Hassard's Life of Arehhishop Hughes. "Wreath of Beauty. Haupt on Bridge Construction. Haven's Where There's a Will There''s a Way. Patient Waiting no Loss. Nothing "Venture Nothing Have. • Out of Debt Out of Danger. Contentment Better than Wealth No Such Word as Fail. All's Not Gold that Glitters. A Place for Everything, and Everything in its Place. Loss and Gain. -: — Pet Bird. Home Series of Juvenile Books. 8 vols, in case. Haven (Memoir of Alice B.), Hazard on the Will. Hecker's Questions of the Soul. Hemans' Poems. 2 vols.' Songs of the AJFections. Henck's Field-Book for Engineers, Henry on HumMi Progress. Herbert's Poems. Here and There. Herodotus, by Johnson (in Greek). Herodotus, by Eawlinson (in English). 4 vols. Heydenreich's German Eeader. Hickok's Rational Cosmology. ■ Rational Psychology. History of the Rebellion, Military and Naval. Illustrated. Hofinian's Poems. Holcombe's. Leading Cases; Law of Dr. and Or. - Letters in Literature. Holly's Country Seats. Holmes' (M. A.) Tempest and Sun- shine. English Orphans. Holmes' (A.) Parties and Principles. Homes of American Authors. Homer's Iliad. Hooker's Complete Works. 2 vols. Hoppin's Notes. Horace, edited by Lincoln. Howitt's Child''s Verse-Book. Juvenile Tales. 14 vols, in case. How's Historical Shakspearian Reader. Shakspearian Eeader. Hue's Tartary and China. Hudson's Life and Adventures. Humboldt's Letters. Himt's (C. H.) Life of Livingston. Hunt's (F. W.) Historical Atlas. Huntington's Lady Alice. Hutton's Mathematics. Huxley's Man's Place in Nature. Origin of Species. Iconographic Encyclopaedia. 6 vols. — 4 Text and 2 Plates. Or, separately : The Countries and Cities of the World. 2 vols. The Navigation of all Ages. 2 vols. The Art of Building in Ancient and Modern Times. 2 vols. The Religions of Ancient and Modem Times. 2 vols. The Fine Arts Illustrated. 2 vols. Technology Illustrated. 2 vols. Internal Revenue Law. Iredell's Life. 2 vols. Italian Comedies. Jacobs' Learning to Spell. The same, in two parts. Jaeger's Class-Book or Zoology. James' (J. A.) Toung Man. James' (H.) L(^ic of Creation. James' (G. P. E.) Adrien. Jameson's (Mrs.) Art Works. Legends of Saints and Martyrs. 2 vols. Legends of the Monastic Orders. Legends of the Madonna. History of Our Lord. 2 vols. Jarvis' Reply to Milner. Jay on American Agriculture. Jeffers on Gunnery. Jeffrey's (F.) Essays. Johnson's M.eaning of Words. Johnson's (Samuel) Easselas. Johnston's CTienustry of Commor Life. 2 vols. Kavan^h's Adele. Beatrice. Daisy Bums. Grace Lee. Madeleine. Nathalie. Rachel Gray. Seven Years. Queen Mab. Women of Christianity. Keats' Poems. D. APPLETON & CO.'S PUBLICATIONS. Keep a Good Heart. Keightley'B Mythology. Keifs Eairy Stories, Keith (Memoir of Caroline P.) Kendrick's Greek Ollendorff. Kenny's Manual of Chess. Kinglake's Crimean War. "Vols. 1 and 2. Kirke White's Poems. Kirkland's Life of Washington. A Cheaper Edition, for Schools. Knowles"* Orlean Lamar. Koeppen'B Middle Ages. Separately— Middle Ages, 2 vols. Atlas. Kohlrausch's History of Germany. Kuhner's Greek Grammar. Lafever's Beauties of Architecture. Lady Alice. Lamarttne's Confidential Disclosures. ■ History of Turkey. 3 vols. Lancelott's Queens of England, and their Times. 2 vols. Landon's (L. E.) Complete Works. Latham's English Language, I^ayard's Nineveh. Illustrated. Cheap edition. Without Illus- trations. Learning to Spell. Le Brun's Telemaque. Lecky's Rise and Influence of nation- alism. 2 vols. Le Sage's Adventures of Gil Bias. 1 vol. Gil Bias, in Spanish. Letter Writer. Letters from Rome. Lewes' (G. H.) History of Philosophy. 2 vols. In 1 vol. Physiology of Common Life. Library of Travel and Adventure. S vols, in case. Library for my Young CountrymeD, 9 vols, in case. Libro Primario de Ortografia, Liebig's Laws of Husbandry. Life of Man Symbolized by the Months of the Year. Light and Darkness Lights and Shadows of liTew York Picture Galleries. Lindsay's Poems. Linn's Life and Services. Little Builder. Little Engineer. Livy, with English Notes. Logan's Chateau Frissac. Looking Glass for the Mind. Lord's Poems. Christ in Hades: a Poem.' Louise. Lunt's Origin of tbe Late War. LycU's Elements of Geology. Lyell's Principles of Geology. Lyra Americana. Lyra Ajiglicana, Macaulay's Essays. 1 vol. Essays. 7 vols. Essays. A New and Revised Edi- tion, on tinted paper. 6 vols. Mackintosh's (Sir J ames) Essays. Madiro. Mahan's Answer to Colenso. Numerals of Scripture. Mahon's England. 2 vols. Maiu's Novum Testamentum Grrece. Mandeville's New Series of Readers. 1. Primary Reader. 2. Second Reader. 8. Third Reader. 4. Fourth Reader, 6. Fifth Reader. Mandeville's Course of Reading. Reading and Oratory. First Spanish Reader. Second Spanish Reader. Third Spanish Reader. Magnall's Historical Questions. Man's Cry and God's Gracious Answer. Manners' At Home and Abroad. Sedgemoor. Manning's Temporal Mission of the Holy Ghost The Reunion of Christendom. Manual of Matrimony. Markham's History of England, Marrayafs Africa. Masterman Keady. Popular Novels. 12 vols. ■ A New and Revised Edition, printed on tinted paper 12 vols. Marryat's Settlers in Canada. Slarshall's (E. C.) Book of Oratory, First Book of Oratory. Marshall's (T. W.) Notes on Episco- pacy. Marshes Double Entry Book-keeping. Single Entry Book-keeping. Bank Book-keepins:. Book-keeping (in Spanish). Blank Books for Double Entry. 6 boolcs in set. Do. for Single Entry. 6 books in sot. Martba's Hooks and Eyes. Martineau's Crofton Boys. Peasant and Prince. Mary Lee. Mary Staunton. Mathews on Whist- Mayhew's Hlustrated Horse Docto May's Bertram Noel. LouivS' School Days. Mortimer's College Life. Sunshine of Greystone. McCormick's Visit to Sebastopol. NJ0M«SlRVA-nON 1996