Small Talk at Wreyland. Second Series

by Cecil Torr

In the Pepysian Library at Cambridge the books are all arranged by sizes; and the arrangement is so rigid that the volumes of a work are separated if there is the slightest difference in their size. But then Mr Pepys had a catalogue of them that was “perfectly alphabeticall.” They are in the bookcases that Sympson made for him in 1666, and they number just three thousand. There is a story that he always kept that number, neither more nor less, turning one book out if he brought another in. But his catalogue has only 2474, and the other 526 were added by his nephew: so it must really be a story of his nephew, not of him.

There was a saying of Mark Pattison’s that no man can respect himself unless he has at least a thousand books, and I have heard it argued that no man need have more. But really it must all depend on what editions they are. There are ninety-four volumes in one edition of Voltaire’s works, and another edition is contained in three. I have these three volumes on my shelves: 6,250 pages with two columns to the page and 78 lines to the column, making about ten million words in all. Goethe’s works are only half that length, but they spread out into five-and-fifty volumes on my shelves: 18,000 pages of 29 lines each.

I have two dictionaries here, written by two old friends of mine—I have known one of them for nearly forty years and the other one for some years more. They both come down to stay with me, but I keep their works apart. Side by side upon a shelf, the dictionaries look like Dignity and Impudence in Landseer’s picture of the dogs. The dictionary of Egyptian Hieroglyphics is the mastiff, and the terrier is the dictionary of Colloquial Chinese. The mastiff is seven times the terrier’s weight and size. But the little one has 1038 pages, of which 1030 are vocabulary, with 45 lines of Chinese type per page. The big one has 1510 pages, of which 1065 are vocabulary, with 60 lines of Hieroglyphic type per page in double column of 30 lines apiece. So the little one is nearly three-quarters the length of the big one, measured in vocabularies, only the paper is much thinner and the type is small—in my eyes, much too small, the Chinese being only a third of the height of the Hieroglyphic, though the characters are more complex.

The author of the Chinese dictionary is also the inventor of a Chinese typewriter. The machine was built by Remingtons; but commercially it has not hitherto been much of a success, as there are no effective laws in China for securing an inventor’s rights. As a rule, it takes two taps to make a character complete, the left half coming with the first tap and the right half with the second. By combining each of the left halves with all the right halves in succession, it can produce a great variety of characters. These, however, are not the characters employed in classic Chinese writing, but more like modern characters evolved from them for ordinary use. Chinese is written in upright lines that are read from top to bottom and follow one another from right to left. Thus, instead of having the characters upright, the typewriter has them lying down. When the paper is taken out, the left edge is treated as the top, and the lines and characters then come in proper order, only the right halves are underneath the left halves instead of being beside them. And this defect could not be cured without making the machine much larger and more difficult to work.

The meaning of the characters varies with the tone in which they are pronounced; and there are similar tones in English. At the tea-party the little girl drawls out, “No, thanks”—tone 1: the hostess thereupon exclaims, “What, none”—tone 2: a friend remonstrates, “Oh, really, now”—tone 3; and the little girl then says what she has heard her father say, “Take the damned thing away”—tone 4. The typewriter puts down the number of the tone, and thus requires three taps in all, two for the character itself and one for this.

This led me into planning a Cuneiform typewriter. I did not have one built, but my plan was briefly this: I divided the wedges that are upright or slope down from the wedges that are level or slope up, and then picked out the commonest groups of each; and, instead of always moving on from left to right, I had a catch to check the automatic movement of the paper, so that one group could be printed on another with the level wedges running across the uprights. The compound groups are not exactly like the groups in printed books; but very often there is just as great a difference between these printed groups and their originals on the clay or stone. After studying Cuneiform in printed books, people are annoyed at finding that they cannot read a word of the real thing; and there are other kinds of print with that same fault.

Hebrew was written in letters that were practically the same as the Phœnician letters from which our English capitals have come: yet it is printed in letters that are not like ours nor like its own. In my Cambridge days the Regius Professor of Hebrew (the late Dr Jarrett) was trying hard to get this changed, and he brought out a Hebrew book in what he called a modified Roman alphabet. I think he might have gone a good deal further, and I do not know that anybody else has gone as far; but, really, something should be done. Hebrew type is just about as bad as any type can be, with half the letters so like others that they may be confused. Greek type is not so bad; but it is quite unlike the Greek of the inscriptions or the older manuscripts. And it expresses nothing that Roman or Italic type would not express as well. If anybody says Iota Subscript, I say iota was not subscript in good Greek, but kept its dressing in the line.

We write English in the same style as Italic type, yet print it with the Roman type that represents handwriting of four hundred years ago. If we are going to copy other writing than our own, we might as well adopt the Gothic type that still survives in Germany. Gothic and Italic both look well, but nothing can be uglier than Roman. I consider it clear proof of mania in a bibliomaniac, if he buys books in Roman type for anything but common use. We ought to scrap the ugly thing, and print English (as we write it) in Italic.

We should be merciful to children. There is quite enough for them to learn, without their learning to read English in one lettering and write it in another. And they might be spared some spelling. Why should they have to ‘proceed’ with e and e together, and ‘recede’ with e and e apart? Both words are based upon the Latin ‘cedere.’ Its participle ‘cessus’ is the base of ‘process’ and of ‘recess’ and also of ‘decease’: yet they may not write ‘decess’ to match, though French has ‘décès’ matching ‘procès.’ Italian always treats the Greek ph as f, and they may do the same in ‘fancy’ and in ‘frenzy,’ but may not do it in ‘philosophy.’ We might at least abolish all anomalies, and also downright blunders like the h in anchor. There are difficulties enough about our spelling without increasing them capriciously.

In early life the mind takes facts in and remembers them, but does not judge them critically, whereas it afterwards becomes more critical and less receptive. It would surely be good policy to feed the mind with facts in the years when it retains them, and leave the reasoning for the years when it can reason. But the policy has been to “make boys think”—at least, that was my experience at Harrow—and that policy defeats itself, as one cannot think effectively without a stock of facts to think about.