19
is an immense number of words in which the senses are different, but the present heterography is the same, and only comparatively very few in which the heterography alters with the signification; fourth, there is a great number of words in the present heterography which have the same spelling but are pronounced differently; or, more properly speaking, there is a great number of different words (different not merely in sense, as in the former case, but in sound, which is the proper characteristic of a word) which are now spelt in the same manner, and these will be discriminated in the new phonotypy in precisely the same manner as they are now distinguished in speaking; fifth, but supposing, even, that every different meaning were distinguished by a different symbolization (which is very, very far, indeed, from being the case, but which must be very nearly the case, in order that the argument to be derived from it should have any weight), this advantage to the eye, when not accompanied by a corresponding advantage to the ear, would be purchased at too dear a rate, as it would so very much increase the difficulty of communicating ideas by writing, on account of the great additional effort of memory which would be required in order to recollect such a large number of different symbolizations of the same sound. Language is a collection of sounds representing ideas; alphabetical writing, a mode of symbolizing the sounds of language. The latter is not responsible for the errors of the former, any more than a mirror is responsible for reflecting disagreeable objects when placed before it. If we wish to have pleasant objects reflected, pleasant objects must be presented; if we wish to have an alphabetically written language, which shall admit of no ambiguities, we must have a spoken one which attains the same object.
We have seen how miserably deficient our present alphabetical system is in the means of representing sounds, and the consequent difficulty of expressing our ideas upon paper; but we must turn to the reverse of the picture. It does not follow, that because a given sound may be expressed in any of 9 or of 20 modes, in the same manner a given combination of letters should represent more than one sound. In fact, we have an instance to the contrary in the French language, in which the difficulty of spelling is, to the full, as great as, or even greater than in English, but the difficulty of reading is very slight; for although each sound is represented by very numerous combinations, each combination represents only one sound, or so very nearly so as to occasion but little trouble to the learner. The second Table, just given, illustrates this inquiry, and the results are nearly equally startling with the former.
We are accustomed to think that the English heterotypic alphabet consists of only 26 letters. True there are only 26 separate different characters, but these are so far from constituting the alphabet of the language, that there are many sounds which cannot be represented by the letters, if we suppose them to have only the values indicated by their alphabetical names. We must, therefore, admit at least, that some or all of the letters should have more than one meaning. But this is not enough; we must allow very frequently that they have no meaning, and also that several when taken together have but a single meaning. Hence in constructing the second table we have taken any combination of the 26 letters to constitute an effective letter of the alphabet, which is on any occasion the
