Almost 250 years ago, in words that better fit our cybernetic age, Jonathan Swift introduced the world to a prototype computer in the middle of Part III of Gulliver's Travels. The Projector-Professor of "the grand Academy of Lagado" had programmed a machine into which he "emptyed the whole Vocabulary. . . and made the strictest Computation of the general Proportion there is in Books between the Numbers of Particles, Nouns, and Verbs, and other Parts of Speech" (GT, 184). Swift has the inventor make the remarkable claim that he can produce great documents of the arts and sciences by even the most random operations of the machine. Significantly enough, the illustration of this table of language is the only non-map plate in the whole book, except for the frontispiece portrait of "Captain Lemuel Gulliver." 18
It is quite possible that Swift himself suggested the form and -- we hope to show -- the model for the illustration of this wonderful machine, since he has Gulliver refer quite specifically to "the Form and Contrivance. . . as in (he Figure here annexed" (GT, 185, our italics). It is, moreover, immensely significant that, to a trained eye, the characters shown in the earliest plate (quite distorted in later issues and editions) -- 256 characters in all -- present a striking resemblance to the ancient Japanese syllabaries which were available to Swift in a special Appendix of Kaempfer's History: a two-page "Alphabeta Japonum."
This discovery, which may be studied by comparison of the plates reproduced here (Plates III and V), has important meaning for both general and specific aspects of research in the Travels as they relate to Japan, the one real "exotic" nation among the imaginary lands visited by Gulliver. In general it confirms our thesis that Swift gave Japan materials great importance in his Travels; Japan is referred to throughout the work and has special importance as the framing at the beginning and end of Part III, where Swift takes pains to show the superior ways of the Japanese people. Specifically, the link of the Travels "figure" and the Japanese language chart as given by Kaempfer serves to confirm overwhelmingly that Swift used the History of Japan whose influences we have seen already in Part II above. There is no other source known to us for the syllabaries of the Japanese language which we can show to have been copied for the language machine illustrated in Swift's Travels. (Although Purchas [III, 466] gives "The Japonian Charter," granted by Ieyasu to the East India Company through John Saris -- a beautifully handwritten document, this is perhaps more difficult to copy than Kaempfer's "Alphabeta Japonum.")
The symbols in the Gulliver "figure" show a high degree of correlation with the Japanese syllabaries as Kaempfer printed them, and -- like the parts of the Projector's "Contrivance" -- the language signs in Japanese are not genuine "alphabet elements" (contrary to Kaempfer's caption) but syllable forms numbering in the hundreds and thus comprising direct word stock for writing. There are three styles of Japanese Kana syllabaries: the Hiragana and Katakana 19 which are still used very much as Kaempfer reproduced them and which seem to dominate in the plate made for Swift's system, and the more complicated Hentaigana which dropped from usage about a century ago (these seldom seen in Swift's "figure"). By close comparison and the compilation of parallels, we find evidence suggesting that Swift and/or the designer of the plate relied heavily upon the table of the Japanese language as given in Kaempfer's History.
To designate the blocks of the 256 characters in Swift's plate, we have labeled the horizontal rows from top to bottom "A, B, C," etc. and the vertical rows from left to right "1, 2, 3,', etc. To the Kaempfer plate we applied a similar system of letters and numbers, prefixing (K) for each such designation. We have tried to decipher each of the symbols in Swift's "figure," comparing it carefully with the script of Kaempfer's "Alphabeta Japonum," and checking our impressions with Japanese scholars and with foreigners who are familiar with the syllabaries. In deciphering these figures, we found it necessary to look at them from every possible angle or position (see Plate IV). Seen upside down, for example, B-1 is the same as (K) C-10: shi G of Kaempfer's Katakana. No two blocks in Travels are the same, showing that the designer was not just scribbling random doodles for Swift but scrupulous in making his symbols and in avoiding repetitions. That they are "Japanese" is no accident.
We have noted two categories, finding at least 46 figures of Swift's table that are clearly legible as single Japanese Kana, and at least 205 more that are composite forms involving two or more Japanese symbols. Only five of the 256 are not reasonably legible (to us and the other Japanese with whom we have checked and compared our observations) as direct elements of the "Alphabeta Japonum."
Using our plates for a block-by-block comparison would go far beyond our limits of space and time (and the reader's patience; we have worked over such detail with Japanese colleagues for years and find an amazing consensus) -- and subjective judgments always are a concern and would make too tidy "conclusions" suspect -- but we think some of the identifications can be accepted by any viewer. Beginning in the right column in the Gulliver's Travels plate (this is the Japanese reading form, as Kaempfer noted in his caption), we find at least nine of the Kana figures suggested clearly in this line alone: A-16 is like (K) I-8 and B-16 is (K) H-14, for example, while composite forms account for most others as in the case of Swift's lower right corner (P-16) which is clearly based on (K) D-8: i. e. to G. Beginning at the top left corner, in A-1 we have the composite of (K) F-16 and (K) G-16 which appear in just the same order and position in Kaempfer as the Travels designer has them, one above the other. We have already noted that B-1 is a Kana figure [(K) C-l0] placed upside down, while C-1 is very close to (K) J-6. Generally, we find that towards the left and the bottom of Swift's plate the designer has moved from his Japanese models as simple forms to combinations of freer improvisation.
Since no other source for the plate of the projector's language machine is known, Kaempfer apparently being the first to present this "alphabet" to English readers, parallels in these plates seem to confirm Barroll's theory that the History was available to Swift and his circle. 20 The consequences of this particular association, as we have seeen, reach far beyond the conclusions that Barroll presented.
In using Kaempfer, not only for this "language machine" but for all the possible borrowings we have noted (acknowledging always as we must that Swift made the most imaginative uses possible of these materials), the incomparable Dean iook his pick of Japan materials and gave them transformed life in his classic Travels. Travelers to Japan, we discover in this new picture of Swift's work that emerges from our findings in Purchas and Kaempfer, provided historical grounding for Part III -- the Japan "book" -- of the Travels but also helped inform the creation of the character of Gulliver himself from beginning to end -- from his acts and condescending interpretations of the land of the little people to his absorption with the life in Houyhnhnmland, where his sense of Occidental superiority was stripped away and, like Will Adams, England's unique expatriate to Japan, he felt no desire to return to the world of barbarian Yahoos.
