You

II.

ON THE STUDY OF LANGUAGE.

ask me whether there is any principle on which a language is to be studied.

And I must own to not a little astonishment at having had the question put to me not by you merely, but by men many years your seniors; because I can't for the life of me conceive how Latin (for instance) is to be taught at all, except on principle.

I will explain my meaning more clearly by illustration. Not long since an excellent mathematical scholar told me that he had been obliged (much against his wishes) to give up Latin and Greek ever since he was fifteen years of age, as he could make nothing of them at all; he much regretted his inability to learn a language grammatically, because (independently of the sources of information from which he was thus debarred) when in India it was of importance to him to know more than one of the native languages thoroughly, but

he never could get beyond the little that his ear taught him a certain conversational fluency, but not one word could he ever learn to write correctly.

On inquiry as to wherein his difficulty consisted, he told me that it arose from no defect of memory; he could easily recollect the meanings of the words in a Latin sentence, but he never could get them into continuous sense without more toil than the results were worth-"le jeu ne vaudrait pas la chandelle "-for, after all, the result was but guesswork.

He used, he assured me (and his plan must seem to any scholar highly ingenious), to work out a sentence on the mathematical rule for permutations and combinations: so many changes of the words were possible—all possible ones must be tried, and the most likely one adopted. Now, I quite believe this to be only an exaggerated form of the difficulties met with by so many in the study of Latin and Greek, for want of having the first principles of those languages placed before them. On going closer into the matter with my mathematical friend, he explained his position more clearly by taking a Virgil and opening it at haphazard; we lighted on

"Res dura et regni novitas me talia cogunt

Moliri, et late fines custode tueri."

Which he considered to mean "he was compelled by the novelty of the state of affairs to commit many cruel actions, and sentinels protected his wide domains;" which, you will see, conveys an idea of the meaning of the passage, but as a grammatical translation was wrong in every particular. I saw at a glance that he had not an idea of the first principles on which the Latin language is based.

Now, all the languages with which you are at all likely to have to do that is to say, the two great languages of the ancient, and two or three of the

modern world-are based each on

ciples, or on a mixture of the two.

one of two prin

There is really

a very broad gulf fixed between the ancient and modern languages: it is simply this, they inflected their substantives, adjectives, and verbs; and we, as a rule, do not. Of all the great modern languages the German inflects most, and the English least.

I will make this distinction, this great gulf of separation, clearer by continuing our conversation relative to these same two lines chosen by the sortes Virgiliana. Where, I asked, in your translation

do you get all your small words from-"he," "was," "by," "to," "of," the plural formations. "sentinels," "domains ?" Oh, all these must be supplied, as they have none of them in Latin. But, excuse me, they have every one of them as rigidly and exactly as we have. Where? I can't see any. Will you be good enough to tell me the meaning of each word, one by one? Certainly: "res " means a "thing, event;" "dura" means "hard;" regni" means a kingdom;" "novitas," "novelty;" "talia," "such;" "cogunt," "compel." Pardon me, you have not given the meaning of any single word right; you have translated only a part of each word. The Latin word for "event" is not "res," but "re;" for "hard" is not "dura," but "dur;" for "kingdom" is "regn;" for "compel" is "cog." You smile at this, but I am not joking; they are simple facts that I am stating. And now, if you want to express the word "event" in the nominative case singular, you say "res," you affix an "s;" if you want to annex to "dur," "hard," the idea of a nominative singular feminine, you affix an a;' "if you want to say "of a kingdom," instead of placing "of" in front, as all the foreign languages except German do, you affix

66

an "i" to "regn;" in fact, "i means "of;" it is no capricious termination; in this declension it can mean nothing else in the singular. Here the nominative of the Latin word "novelty" is needed; in more grammatical language you want to employ "novelty" as the subject of the sentence, so as" is added to "novit;" so "talia" is not "such," but "such things," not "such men or women," nor by" nor "with" nor "from" nor "of" nor "to such things," but simply "such things," nominative or accusative; so "cogunt" is not "compel," that is not enough, but "they are compelling," or "they compel;" "cog" equals "compel," "unt" equals "they." This is the principle, the very opposite of our modern languages' principle, and it is carried out to a much greater extent in Greek than in Latin. Once master this idea; look carefully for all the small words in the right place, namely at the end of the rootwords, and I will venture to assert that, with your mathematical memory, you will not be long before you read Latin nearly as easily as you read the Times. But would you teach languages that way? Undoubtedly, as soon as the rudiments of the grammar are mastered: you must know the Latin