By this time we have fairly encroached upon the question as to the method to be used in training pupils in the grammar of a foreign language. I want to introduce my discussion of this subject with the following quotation from N. M. Petersen (Sprogkundskab i Norden, Collected Works, Copenhagen, 1870, ii. 297–8):

“With respect to method, the artificial one must be given up and a more natural one must take its place. According to the artificial method, the first thing done is to hand the boy a grammar and cram it into him piece by piece, for everything is in pieces; he is filled with paradigms which have no connection with each other or with anything else in the world ... he is filled with words, only half of which occur occasionally, and some never at all in what he reads. How old are not the complaints over this perverted method! how many sighs it has occasioned, how much deformity it has produced! On the other hand, the natural method of learning languages is by practice. That is the way one’s native language is acquired. The pupil becomes acquainted with the elements and absorbs them, as it were, into his soul in their entirety before he is consciously able to separate and account for the single parts and their special relations; he forms whole complete sentences without knowing which is the subject and which the object; he gradually finds out that he has to give each part of the sentence its correct endings without knowing anything about tense or case.... The logical consequence of this, then, is that as a rule one cannot begin with grammar in teaching languages to a child of ten or twelve. His first years at school ought to give him merely materials; he ought to collect experiences (that is a child’s greatest delight), but not speculate over them.”

It is now half a century ago since N. M. Petersen uttered these golden words, and still the old grammar-instruction lives and flourishes with its rigmaroles and rules and exceptions, that intensely stupid custom, the teaching of grammar to children, as Herbert Spencer calls it. Only few of the boys in our schools who have studied German for several years, are able to connect for instance um with the proper case without hesitation; but there are certainly still fewer who cannot run through durch fÜr gegen ohne um and wider like parrots. But strangely enough this ever present phenomenon does not yet seem to have led to a general acknowledgment of the fact that these grammatical rigmaroles as a rule are scarcely worth as much as the counting-out rigmaroles of the children: eeny meeny miny mo.[19]

And, of course, paradigms which are learned by rote also belong to the category of rigmaroles. “Paradigms ought by all means to be given, but should never be learned by heart in rigmarole-fashion.” (N. M. Petersen.) Thoughtlessness and stupidity thrive excellently on this continual repetition of words as words, that is words without any mutual association, without connection in sentences. Just think of the many thousands of boys and girls who time and again recite: mourir, mourant, mort, je meurs, je mourus, and then ask how many of them, yes even of their teachers, ever happen to think that the last form in reality is impossible (at all events in conversations in this life).[20] The percentage is scarcely very large. And when conscientious philologists like Ayer and Sachs give imperative forms like nais, naissons, naissez—be born! let us be born!! be ye born!!! it cannot be denied that we are tempted to use the exclamation: “die gelehrten, die verkehrten!” Of course it is not our aim to get rid of such forms as je mourus;[21] what is wrong is the system. I condemn vivre, vivant, vÉcu, je vis, je vÉcus just as strongly as mourir, etc., even if none of these forms is really meaningless. And the reason why I reject this method of teaching languages is because it does not and cannot bring us to our desired goal. The chief absurdity, the one which it is our business to quarrel with, is that use of disconnected words for grammatical purposes, which flourishes in all our text-books.

It has often amused me to examine grown-up persons (non-philologists) in what they could remember of the instruction they had received in school in foreign languages. It seems to be extremely common that they have not the slightest idea as to what case for instance a preposition governs, but the rigmarole in which it occurs they generally know by heart. They also know ever so many scraps like der buchstabe, der friede, der funke ... or das amt, das ass, das bad, das bild, das blatt ... but why they have learned these things, and what they were supposed to be good for, to these questions there is generally no answer forthcoming. So those rigmaroles are really of no practical use whatever.

Now, of course, rigmaroles could easily be so arranged—though no one seems to have put it into practice—as to contain an indication of the object in grouping together just those words, for instance by saying durch das zimmer, fÜr, gegen ... or durch fÜr ... um wider mich, or das amt, die Ämter, das ass ... or das amt, Ämter, bÄder, bilder....

But even in this improved form it seems to me that grammatical rigmaroles are of little value just because they accustom the pupils to learn and say things by rote without thinking; they are remnants of the old-fashioned would-be pedagogy where a teacher in any subject was satisfied if the pupil only “knew his lesson,” that is, could recite the words of the book, and where no one ever thought about understanding or other such-like modern inventions.

The expressions “living” and “dead” are so often used about languages and words, but those who use them do not always take the trouble to consider in what sense these expressions really have any meaning. A language only lives, and can only live, in a person’s mind, and that it lives there means that its component parts are for him associated with certain ideas, which are recalled when he hears the words, and which in turn summon up the corresponding words when he wants to express them, or when he simply wants to make them clear for himself. But ideas do not and cannot exist except in combinations; an absolutely isolated thought is the same as nothing. It is the same with words; if they are taken out of their natural surroundings, they suffer atrophy and at last cease to perform the usual function of words, namely to produce ideas. So isolated words, such as are given in rigmaroles and paradigms, are only ghosts or corpses of words. Try to run through the words “jewel, stone, cabbage, knee, owl, toys, louse,” and see if a single complete picture has been produced in your mind—but you are no better off when you say the French rigmarole bijou, caillou, chou, genou, hibou, joujou, pou. That, as well as amo, amas, amat, amamus, amatis, amant and all the others, must by virtue of the fundamental psychical law of the life of language become merely empty jingle and nothing else. Now we see the psychological reason why sensible persons can write such sentences in their books as je mourus or the entirely parallel “Wir sind nicht hier.” When the mind is occupied with a word as a grammatical phenomenon, the word’s normal power of calling forth ideas is of course lessened in a considerable degree.

Furthermore the isolation of words for grammatical purposes may even lead us to make positive mistakes. The pupils are first carefully taught in the grammar that “nobody” in French is ne personne and “never” ne jamais,[22] and later on it is corrected as a serious mistake when they write ne personne parlait or il ne jamais parle, mistakes which would never have occurred if the pupils had not been allowed to learn the false formulation. In modern French “nobody” is personne and “never” jamais, just as “not” is pas, etc. Ne only exists in connection with a verb, and ought never to be seen or learned by the pupils except in its natural surroundings; out of connection it is no more a word than un (in unfriendly, ungracious, etc.). The rule for its employment can be thus stated in short, that it is placed in front of the verb, always, if the sentence is wholly negative, also often if it is only half negative (by which I mean the well-known cases after empÊcher, craindre, comparatives, etc., where ne is well on the way to slip out of the living French language, and where we now, after the last ministerial decrees, may allow ourselves a little laxity in teaching these points).[23] Likewise it is only injurious to teach the children that “I” is je, “thou” tu—a matter of fact it is moi, toi, while of course “I go, thou goest,” is je vais, tu vas; what usage has joined together, let no grammar put asunder.

But words, when in their natural connections, show their vitality in other ways besides in summing up the correct ideas; they have another power, which they also lose when they are isolated, namely the power of breeding new connections in the image of the old ones. If I have often reproduced a certain type of word-formation or sentence-construction, then this becomes a part of my mental mechanism in such a way that I unconsciously make something new (coin a new word, construct a new sentence) after the same pattern, after the “analogy” of what I know, whenever I need it, just as the English boy who has often heard superlatives like hardest, cleanest, highest, etc., does not need any rule to be able to construct forms like purest, ugliest, dirtiest, of his own accord, and who, at the moment when he says them, would not be able even by means of the most scrupulous analysis to decide if he has heard the form often before and is merely reproducing it, or if he himself is creating it without having previously heard it—and, if the latter is the case, if he is creating something which others also have created, or if it is the very first time that the word is used in the language—this is what takes place every minute wherever human languages are spoken.[24] An Englishman has so often heard (and repeated) sentences like “give the man your hand,” “I gave the boy a whipping,” “he gave his sister an apple,” that he unconsciously forms his sentences according to a scheme where the indirect object always precedes the direct object, and which even without this grammatical terminology and without any rule would lead him quite naturally to say, for instance, “Will you give your father the money?” A Frenchman would just as instinctively say, “Veux-tu donner cet argent À ton pÈre?” because in all the sentences which he has experienced he has heard the “dative” expressed by À after the direct object.

But since this takes place by virtue of inviolable psychical laws, it applies not only to the mother-tongue, but also to the foreign languages which we learn later. We simply cannot avoid thus unconsciously forming types or patterns to go by, in using a foreign language, as soon as the conditions for these typical formations are at hand. If, on learning English, a Dane has frequently heard (read) and (especially) used combinations like up here, in here, in there, out there, then he will quite naturally say down there when he wants to express this thought; it is not at all necessary for him previously to have learned a rule to the effect that “here and there in connection with other adverbs of place stand last.” As a matter of fact, when we speak or write a foreign language, we employ a number of such rules which we have never seen formulated, and, what is more, also rules which have never at any time been consciously formulated by any grammarian. The reason why we cannot attain the same confidence in all departments of the foreign language that we feel in our native language is of course partly because the conditions are not so favourable, and partly because our mother-tongue acts as a hindrance on account of the tendency it has to intrude on all occasions and mislead us to construct sentences after its pattern.

But the conditions become the more favourable for this unconscious mental activity in our pupils the more we know how to make each sentence in the foreign language have its full effect upon them and become their possession, and the more we can keep the mother-tongue in the background. And although we can never bring it about that our pupils come across the forms in the foreign language even approximately as often as that child does who is learning his native language, yet we can to a large extent make amends for this by bringing a better system into our teaching, so that the acquiring of the language will not depend so much upon chance as is the case when babies learn to talk, just as it is also an advantage that our pupils are older and more developed, and that we can get some help from the written and printed language.

Many of the transposition exercises mentioned in the last section are essentially grammatical, but we can easily hit upon still more exercises by which we may in a systematic way encourage the natural tendency toward type- and series-formations. To conjugate a verb all the way through by itself is the sheerest drudgery, but the exercise immediately becomes both more interesting and more beneficial when it is a whole sentence that is to be tackled. For instance, the teacher can write on the blackboard a sentence like “Je donne un sou À Alfred” and get the pupils to conjugate it through all the persons. In the beginning he might also write down all the forms of the verb, one under the other; they are not to be committed to memory, but merely furnish a scheme, which the pupils are to fill out by inserting the correct pronouns before, and un sou À Alfred after the verb. Then the next step is to let the pupils use other words instead of un sou and Alfred, so that pupil A says, for instance, Je donne un centime À Paul. B: tu donnes un franc À Jean. C: il donne un livre À papa. D: nous donnons des poires À l’Épicier, etc. Then in reality the task which the boys have before them is to hit upon new words to insert (they must make sense!); consequently it becomes a kind of game in which the vocabulary is reviewed like the one mentioned above (p. 99), but at the same time the forms of the verb are practised. If a pupil should happen to say, for instance, ils donnent deux cerises À le maÎtre, the teacher must only say the sentence himself with the correct au and make him repeat it in this form without scolding him,—yes, even without stopping to give a long explanation of why it should be au and not À le in this case. This kind of exercise can of course be varied in different ways; such a sentence as mon pÈre me donne de l’argent is written down, and the pupils are told to inflect it in all the persons, which of course only involves an alteration of mon and me; or the sentence is to be reconstructed with other tenses, etc. More complicated sentences, too, may be conjugated all the way through, either without changing anything but the pronouns and the forms of the verbs, as for instance, Je suis allÉ me promener avec mon pÈre; Das habe ich ihm gestern versprochen, und ich werde es ihm morgen geben—or in such a way that other things are changed too: je m’appelle ... where the pupil is to insert real names (his own, a comrade’s ... in case it is vous, the teacher’s); Ich habe meinen vater um etwas brot gebeten. Du hast deinen vater um etwas geld gebeten. Er hat seinen vater um ein stÜck papier gebeten. Sie hat ihren vater um einen kuchen gebeten, etc. Of course one can also assign written exercises of a similar kind, as for instance: construct five sentences like Le pÈre de Jean est allÉ À la maison de sa soeur, using different words in each sentence in place of those here italicized, etc., etc.; but it were best if these sentences were suggested by, or in some way associated with, sentences in the text-book.

Now some people will say that this is only another way of employing those grammatical isolated sentences which I have declaimed against—and they are right in so far as I admit that the more the exercises are made to resemble the old-fashioned ones, the poorer they are for the purpose, and if they are employed to too great an extent they may easily degenerate into tiresome mechanical routine-work. But if used to moderation they will only be beneficial, and then, besides, they differ from the single sentences of the old method in being associated with a text which has been read, so they are not thus quite isolated from a sensible connection; they also differ because translation is not used and is not needed (except when the teacher at long intervals has to make sure that pupil A has understood a sentence given by pupil C, who has used an unusual word); they differ because, translation being omitted, the whole exercise can proceed at a rapid pace; they differ because the sentences are constructed by the pupils themselves, who are thus compelled all the time to pay attention both to their form and contents; and finally they differ because, as a result of all this, they are more interesting and amusing to the pupils. Furthermore such exercises incite the pupils to want to say something of their own accord, and thus they get a desire to extend their knowledge; they will frequently ask what this or that word which they need in a sentence is in French or German—and in that case the teacher must always answer, but then he must always require, too, that they learn the word which has been given them (to prevent them from getting into the habit of asking superficially and carelessly just “for the fun of it”). Finally the pupils will thus be brought to appreciate the benefit of learning grammar; their grammatical knowledge is not sheer theory for them, but is continually converted into effective power and thus becomes easier to remember, for there is no doubt that Goethe is right when he says: “Still all that we can remember of our studies in the end is only what we have been able to find practical use for.”

Of course, the sentences constructed by the pupils in the course of any one of the exercises recommended in this book may contain mistakes, and the most serious mistakes must be corrected, yet with as little particularity as possible, if they have nothing to do with the phenomenon which is being or just has been carefully considered and practised, and with as few theoretical reasons as possible. Many exercises can be so arranged that it is scarcely possible for the pupils to make any mistake, and this without becoming less valuable; on the contrary, they will often be the best, for every sentence which a pupil constructs or says correctly confirms good habits of language. But no matter how much one may favour the theory that “Prevention is better than cure,” it is not well to be too anxious to prevent mistakes. One of the ablest advocates of the reform in Germany, Wendt, says: “It is of more importance for the pupil to talk at all than to talk correctly,” and although I know what criticism I have to expect from unsympathetic opponents about my encouraging superficiality and not caring a bit about correctness, yet I cannot deny myself the pleasure of quoting with approbation a Slavic proverb, Tko zeli dobro govoriti mora natucati (whoever wants to speak well must murder the language), which Schuchardt has chosen as a motto for his stimulating work about mixed languages,[25] and which he interprets: “Wer aus irgend einem grunde sich scheut eine fremde sprache zu misshandeln, der werd sie nie beherrschen.”

In order to reassure people who cannot help feeling anxious, I shall add here three statements from the report of the ninth German “Neuphilologentag” (1901). Klinghardt (p. 100) confesses that he has been converted to the reform, because, in spite of years of vigorous efforts, he had not succeeded by means of the translation method[26] in training the majority of his pupils to grammatical correctness. Headmasters of schools where the old method was employed had also told him that there were still serious grammatical mistakes of form in the written exercises which were handed in at the final examinations. But, after he had given up the translation procedure, all of his pupils, even the backward ones, had attained to grammatical correctness. Wendt (p. 101) emphatically denied that anything could be gained in grammatical sureness by translation exercises. And Walter (p. 102) repudiated the accusation which is always on the tongue of many of the opponents of reform, that the reformers entirely do away with grammar, by referring to many of these very gentlemen, who, on visiting his school, had expressed surprise at the grammatical sureness displayed by his pupils.

And since I now seem to be in the mood for quotations, I can also refer to Goethe’s words: “Thus I had learned Latin, just like German, French, English, only through practice, without rule and without system. Anyone who knows what the state of school instruction was at that time will not find it strange that I neglected the grammar as well as the rhetoric; everything seemed to come naturally to me. I retained the words, their formations and transformations in my ear and in my mind, and I employed the language with ease for writing and talking.”[27]

In giving the pupil English sentences to translate into the foreign language, we are only artificially creating difficulties. If it is difficult for the pupil to translate into his mother-tongue where at least confirmed habit ought to prevent him from falling into the worst pitfalls, then it must be much more difficult, indeed impossible, to translate into a foreign language where he is not yet quite at home. We ourselves lead the pupil to make mistakes, and then we have to do all we can to prevent his confronting us with a too overwhelming number of them. To this end we limit each exercise to illustrating one, or two, or three, paragraphs in the grammar; we make theoretical rules to serve as a guide in translating, without always remembering how difficult it is to make practical use of such rules; we bracket the words which are not to be translated; we try to be helpful by placing alongside of, or underneath, the correct English, some very strange English indeed, which, however, has the advantage that it can be translated literally, etc., etc. And the result of all this exertion? Well, it is a well known fact that they are not always things of beauty that we meet with in the French exercises which are handed in after many years of toil, according to this method. Experience is sure to teach us that this is not the means to our end. Joh. Storm is right when he says (Franske taleÖvelser, Preface): “The worst and most unfruitful torment in the school instruction of the present time is the excessive use of written exercises in foreign languages.” As a bright contrast to this “constructive” method of procedure, we have the “imitative” method, which may be so called partly because it is an imitation of the way in which a child learns his native language, partly because it depends upon that invaluable faculty, the natural imitative instinct of the pupils, to give them the proper linguistic feeling, if it only has ample opportunity to come into play. As a motto for this method, we might perhaps say: Away with lists and rules. Practise what is right again and again!