Gradation means passing from the known to the unknown by easy stages, each of which serves as a preparation for the next. If a student who is willing to learn and is capable of learning finds his lessons too difficult, if he fails to understand or to apply correctly the explanations we give him, if his rate of progress is too slow, if he forgets frequently what he has already learnt, and if his oral or written work is characterized by an excessive degree of inaccuracy, it is perfectly certain that his course and his lessons are badly graded.

The student’s progress may in the initial stage be slow; after ten or twenty lessons he may not seem to have advanced very far; but if he has been laying a good foundation he has been doing good work, for it will mean that the next stage of his work will be accomplished more easily and more rapidly. During the first lessons he is not so much learning the language as learning how to learn it. During the second period his progress will be more rapid and he will assimilate more of the actual language-material, and he will then be learning in such a way that the third stage will be still more rapid, and so on through the subsequent stages; his rate of progress will increase in proportion as he advances.

In the ideal course, this principle will be observed in the fullest possible measure; the course itself will be divided into appropriate stages, each of which will be marked by an increased capacity on the part of the student for assimilating and using language-material.

The vocabulary in a well-graded language-course will be arranged in such a manner that the more useful words will be learnt before the less useful. (Let us remember that there are two sorts of ‘useful words’: those which are useful in themselves on account of their intrinsic meaning, and words which are useful as sentence-formers.) The rate of progress on the part of the student will depend very largely on the manner in which the vocabulary is graded. Twenty-five well-chosen words will form more useful sentences than many people believe; with five hundred well-chosen words an incredibly large number of valuable sentences can be formed. For detailed information on this point we would refer the reader to the statistics which have been compiled by those who have made a special study of this particular subject. In the ideally graded course the student first assimilates a relatively small but exceedingly important vocabulary; he learns to use it, he learns the more important peculiarities of each word, he learns how to combine these words in sentences, he learns the exact range of meanings covered by each word either singly or in combination with its fellows. This small vocabulary then constitutes a sound nucleus, and this nucleus is of twofold utility; it not only provides the student with useful words, with language-material which he can actually use, but it serves at the same time as a sort of centre of attraction for new language-material. The most apt illustration of this form of gradation is the snowball, the huge mass of snow which accumulates rapidly and easily once we have provided the nucleus represented by the first compact and well-rounded handful.

The grammatical material must also be graded. Certain moods and tenses are more useful than others; let us therefore concentrate on the useful ones first. In a language possessing a number of cases, we will not learn off the whole set of prepositions, their uses and requirements, but we will select them in accordance with their degree of importance. As for lists of rules and exceptions, if we learn them at all we will learn them in strict order of necessity. In most languages we shall probably find certain fundamental laws of grammar and syntax upon which the whole structure of the language depends; if our course is to comprise the conscious study of the mechanism of a given language, then, in accordance with the principle of gradation, let us first learn these essentials and leave the details to a later stage.

Gradation can and must also be observed in the study of the semantic aspect of a language. If a given word has several meanings, let us first associate the word with its more usual or useful meanings. If a foreigner is to learn the English verb to afford, let him begin by using it in such sentences as I can’t afford it, and not in such examples as It affords me the necessary opportunity. If we are teaching French, let us first use ciel in the sense of sky and leave the idea of heaven to a later stage. I may go has more than one meaning, but let us first teach it in the sense of Perhaps I shall go; the other varieties are not of pressing importance. When we introduce but, let us associate it with its usual meaning and forget for the moment that semantic variety which is equivalent to except.

It is not sufficient for us to adopt the general principle of gradation; we must adopt the right sort of gradation; for we can easily imagine all sorts of false grading. We can imagine a teacher refusing, on the score of gradation, to teach irregular forms before regular ones, and justifying his procedure by the assertion that the regular is easy and the irregular difficult. This kind of gradation, however, is obviously unsound, seeing that some of the most useful words in most languages are very irregular. As a matter of fact, in a sound course of study based on the principle of automatism the irregular forms are learnt as easily as (and sometimes more easily than) regular forms.

We can also imagine some teachers maintaining (on grounds of gradation) that the word should be treated before the sentence. They would say that it is easier to assimilate a word than a sentence, that what is easier should come first. Others might say, “Teach easy words first and difficult ones later.” But this cannot be right, for if we observed this rule we should teach a Spaniard or a Frenchman the English verb to comprehend before the verb to understand. Gradation does not necessarily imply passing from the easy to the difficult, but it always does imply passing from the more important, useful, or fundamental to the less important, useful, or fundamental. Now, whatever the true unit of speech may be, our leading semanticians and speech-psychologists are all agreed that this unit is rarely the word, but generally the word-group or sentence. Consequently, to start from the word is not only bad gradation but bad semantics.

We have heard it asserted on grounds of gradation that the written form of a language should be studied before its spoken form. Here again we find the same misinterpretation of the term gradation and the same fallacy of ‘facility.’ If relative facility is to be the basis of gradation, then we should teach the geography of Portugal before teaching the geography of the British Empire, and we should postpone our study of chess until we have become expert in the easier game of noughts and crosses. To learn how to read and to write a language may possibly be easier than to learn how to speak it and to understand it when spoken, but this has no bearing on the subject of gradation.

Another false interpretation of the principle is to assume that the student will begin by using incorrect or pidgin-French (or whatever the language may be) and will gradually become more perfect ‘with practice’ as he goes on. Now if this is gradation, it is a particularly vicious form, and in flat contradiction to the principles of habit-forming and accuracy.

To teach to adults ‘child-like words’ before the words used by adults is another misinterpretation of gradation. We do indeed see classes in which boys or girls of twelve or thirteen learn to recite foreign nursery rhymes, but we doubt whether any teacher would seriously maintain that words such as dog or sheep should be preceded by bow-wow or baa-baa.

Having examined some faulty and vicious manners of interpreting the term gradation, let us now proceed to epitomize a few rational applications of the principle we have set forth, and let us assure the reader that each one of these has been proved to be psychologically sound.

(a) Ears before Eyes.—All fresh language-material should be presented in its oral form and not in its written form. Sounds should first be practised without any reference to any graphic forms of representation; the ear, not the eye, is the organ provided by nature for recognizing and assimilating sounds. Words should first be heard and imitated orally, for ideal assimilation is not helped but hindered when the written form is present.

Fresh word-groups and sentences should also as far as possible be first introduced and learnt orally. The adult student who complains that the process is too difficult is under the illusion that we hear with our eyes.

(b) Reception before Production.—It is quite certain that the student will be unable to reproduce a sound, a word, or a word-group that has been pronounced to him until he has really heard the model that he is called upon to imitate. There is a great difference between really hearing and merely imagining that one has heard a sound or a succession of sounds. As a rule we do not hear what is actually said to us; we merely hear what we expect to hear. Ask the average foreigner to repeat after you a word such as und’stand; instead of reproducing the exaggeratedly shortened form as represented above in two syllables, he will say understand in three syllables. As a matter of fact, he is under the impression that he heard you articulate the three syllables, and consequently he reproduces what he thought you said. Ask the average foreigner to repeat after you the word turn pronounced in Southern English (i.e. without an r) and he will insert an r because he imagines that he heard one. The sentence il doit venir is pronounced by the average Frenchman as [idwavni?r]; pronounce it like that to the average English student of French and ask him to imitate you; in most cases you will obtain what the student imagined he heard, viz. [il dwa v?ni?] (the last word having a remarkable resemblance to the English word veneer).

The student must therefore not only be trained to hear, but in all fairness to him he should be given ample opportunity of hearing the sound, word, or word-group that he will be called upon to reproduce. Let him hear it several times, let him concentrate his attention on the succession of sounds without any regard to its written form or its meaning. Let us endeavour as far as possible to give the student two or even more separate opportunities (with appropriate intervals) of truly hearing any given sound, word, or word-group before calling upon him to imitate the model.

(c) Oral Repetition before Reading.—Just as oral repetition should be preceded by a period of audition, so should reading be preceded by oral repetition. Before calling upon a pupil to read off from the blackboard or his book a word, list of words, sentence, or succession of sentences, let us first ask him to repeat after us the required material. If he cannot reproduce to our satisfaction a sentence that he has just heard from our lips, he will certainly be unable to reproduce the sentence by the process of reading.

(d) Immediate Memory before Prolonged Memory.—The teacher pronounces a sound or a succession of sounds (a word, a word-group, or a sentence). A few seconds later the pupil reproduces what he has heard; he does not find it very difficult to do so, for the sound of the teacher’s voice is still ringing in his ears; in his imagination he can still hear the teacher’s voice, and he has but to speak in unison with it. The sort of memory which enables him to reproduce what he has just heard is called immediate memory.

Another time the teacher pronounces a sound or a succession of sounds. The next day the pupil is called upon (without being prompted) to reproduce what he heard the day before. He may fail altogether to do so, or he may succeed. That sort of memory which enables him to reproduce what he has heard one, several, or many days before is called prolonged memory. Let us be quite certain that we understand the difference between these two extreme varieties of memory. Let us choose a word in a language unknown to us. Let this be the Hungarian word szenvedni, meaning to suffer. The word is pronounced [´s?nv?dni]; in the absence of a teacher let us pronounce the word ourselves three or four times.... Let us take our eyes from the book, and let a few seconds pass.... How do you say to suffer in Hungarian? [´s?nv?dni.] Quite correct; we have reproduced the word from immediate memory. To-morrow or the day after let us ask ourselves how they express to suffer in Hungarian. If we are able to answer correctly without referring to the book, it will be by dint of our prolonged memory. If this experiment is inconclusive, let us take a word-group, a sentence, or a list of words; we shall then realize how much more difficult it is to reproduce new matter after the sound of it has faded from our ears.

Let us remember this experiment when we are teaching; when our pupils reproduce correctly (either by repetition or by translation) what they have heard a few seconds (or even a few minutes) before, let us remember that we have so far only appealed to their immediate memory, and let us not expect an equally satisfactory result when we call upon them to reproduce the same matter the next day without prompting. To expect the same results from the prolonged as from the immediate memory implies a faulty grasp of the principle of gradation. Let us give our pupils ample opportunities, on an appropriate number of occasions, of reproducing matter heard a few moments previously; this will strengthen their associations and when later on we appeal to their prolonged memory, the results will be satisfactory.

(e) Chorus-work before Individual Work.—Before we call on an individual pupil to articulate a sound or a succession of sounds, let the work be done in chorus on two or more different occasions. For an individual to have to submit his tentative efforts to the criticism and perhaps laughter of his fellow-pupils is not conducive to good results. Let the individual pupils test their articulation in company with others, and when by so doing they have gained a certain mastery of what they have to repeat and have thereby gained a certain degree of confidence, let them proceed to reproduce singly what they have previously phonated together.

(f) Drill-work before Free Work.—This is perhaps the most important of the precepts to be observed in connexion with gradation. The forms of exercise to which the general term drill-work may be applied are many and varied. Some of them are calculated to train the student in perceiving and discriminating the sounds of which the language is composed; others are articulation exercises; there are also special forms of drill-work which aim at securing fluency and accuracy in producing successions of syllables. The question-and-answer method may be embodied in many interesting forms of drill-work; there exist also many varieties of action drill, conversion drill, translation drill, and grammar drill. All these forms are characterized by common attributes: they are all systematic, highly organized, and susceptible of infinite gradation; the work is methodical and proceeds steadily and continuously without breaks or interruptions. Most forms of drill-work have been composed and are carried out in such a way as to preclude the possibility of the student’s forming bad habits. Indeed, if the work is carried out as designed the element of error should be almost entirely excluded.

Now free work in all its varied forms, such as free conversation, free translation, and free composition, differs greatly from drill-work, and we can all testify to the ludicrous results these forms of work yield when performed by one who has had no previous drilling. If the student has not been put through a proper course of drill-work, all his efforts at free work will be based on that most unnatural and vicious of processes—conventional translation from the mother-tongue. The undrilled French student will be speaking and writing not English as we understand the term, but anglicized French. Having formed no English language habits, he will cast all his thoughts in the French mould, and when the exact English equivalents to his French words and phrases are missing he will break down.

Free work without the essential preparation means faulty work, uncertain and erratic work; it means the formation of nearly all the bad habits which characterize the average student and which mar his work.