Another way of learning the letters is given on a subsequent page (the 79th); but this has the advantage of being a little more scientific, and exercising the classifying instinct, which has been considerably developed by the exercises involved in the occupations.
On account of the irregularity of what is called English orthoepy and orthography, the written language is a chaos—into which, when the child's mind is introduced in the usual way, all its natural attempts at classification are baffled. The late Horace Mann, in a lecture on the alphabet, has with great humor and perspicacity shown this; and he recommended that children should be taught to read by words purely. But when some years afterwards his attention was drawn to the phonic method, he accepted it fully; and wrote for Mrs. Mann the preface to her Philadelphia edition of the Primer of Reading and Drawing. This was not until after it had been tested in his own family and some others, where I had introduced the phonic method.
On the details of my method I must enlarge all the more, because I find myself differing in some respects from Mr. Sheldon's plan, which loses a large part of the advantages of the phonic method by not having one definite sound for each letter. As I have taught on my plan successfully for fifteen years, I am prepared to defend it at all points, from the ground of a various experience. But I can adduce also the highest philologic authority for my mode of sounding the alphabet,[E] as well as an argument of common sense from the nature of the case.
The primal cause of the chaotic condition of English orthography, is the fact that the Roman alphabet, which was a perfect phonography of the old Latin language, lacked characters for four English vowels and four English consonants. The Latin monks had not the wit to invent new characters for these additional sounds; but undertook to use the Roman letters for them also. Hence for the vowel heard in the words irk, err, work, and urge, they used indifferently all four characters; for truly one would do as well as another. But if they had put a dot into the middle of the o, and added it to the alphabet, it would have been better than either. Also, if for the vowel sound of pun, they had put a dot under the u; and for the vowel sound of man, they had put a dot under the a; and for the vowel sound of not, a dot under the o; they would have had four more letters in their alphabet, which would have completed the phonography of the English vowels. Similar dots under d t s c would have made a phonography of consonants, and avoided the awkward combinations of sh, ch, and the ambiguity of th, which now stands for the differing initials of then and thin.
But as they did not do this, a certain divorce took place between the ideas of the sounds and the letters; and hence the long uncertainty of the English orthography, and the stereotyped absurdities which now mark it.
It is so nearly impossible to remedy a difficulty which has passed into print so largely, that we have to accept the evil, and remedy as best we may the disadvantage it is to young minds to have all this confusion presented to them on the threshold of their literary education.[F]
It was suggested to me by Dr. Kraitsir, that I should take a volume of any book, and count the times that each of the vowels, and c and g, were sounded as the Romans sounded them, and how many times they were sounded otherwise, and thus see whether it was true, as he said, that these Roman sounds were the most frequent, even in the English language. I did so on a few pages of Sir Walter Scott's novels, and found that the letter i sounded as in ink 240 times, to one that it sounded as in bind; and though the proportion was not quite so great with any other vowel, yet there was a large majority for the Roman sound, in each instance, as well as for the hard sounds of c and g. Indeed I found g was hard, even before e and i, in the case of every Saxon word; and that all the soft gs, which are not many, were derived from the Norman-French.
I then set myself to find what words in English were written entirely with the Roman-sounding letters; and, to my surprise, found a large number,—enough to fill a primary spelling-book;—while most of the syllables of the rest of the words in the language yielded on analysis the same sounds. It immediately occurred to me to begin to teach children to read by these words, whose analysis would always yield them the Roman sounds, and reserve, till afterwards, the words which are exceptions, leaving the anomalies to be learnt by rote.
I tried my first experiment on a child a little more than four years old, by printing on a black-board certain words, letter by letter, until he had learned the whole alphabet, both to know each character at sight, and to print it on the black-board, and it was a signal success.