Not all of the keys on this keyboard produce letters. Obviously there are control characters but leaving those aside, if you were to read some of the characters on here, even if you knew no English, if you have been able to hear, a voice would pop up in your own head which would say them in your own language. For instance, I could type “1984”, and the chances are you would hear something like “neunzehnhundertvierundachtzig” or “dix neuf quatre vingt quatre” (sorry about the inverted commas). When I typed that, I thought the first thing but you may have thought “nineteen eighty-four” for example. There is a sense in which these characters transcend language but continue to represent concepts, which are expressed differently in different languages. Just to doodle vaguely on the keyboard, I can produce a list of such signs with ideographic rather than alphabetic meaning, thus:
¦, ¬,£,$,€,&,*,-,+,=,@,#,~,/.
Not all of these are going to mean something to everyone who reads them, and some of them will mean substantially different things, but I read the above list as follows:
Not both . . . and . . ., not, pounds, dollars, euros, and, multiplied by, minus, plus, equals, at, number, roughly, divided by.
A couple of the signs have more than one meaning and for me there’s a substantial number of other signs, widely used in technical contexts, which still have meaning for me but probably not for most people:
℞, c̄, →, ↔, ∨, ⊕, ∀,∃,⊢, ⟡, □.
These are: prescription (I think of it as a Demotic eye of Horus but that may be incorrect), with, materially entails that, if and only if, and/or, either . . . or . . . , for all of, there exists, it is a theorem that, contingently, necessarily.
There’s a third set I don’t know how to type on a computer which is taken from Blissymbols, a system of signs used for people with communication difficulties, but which seems to have been initially invented by Charles Bliss in 1949 as an international language. Since I use these only in writing, I have no idea if they’re in the Unicode character set, and it would defeat the object of typing or writing less if they were because it would take something like eight keystrokes to type the character which means “see” for example, and having just tried to type that it turns out this computer doesn’t have any in its character set, annoyingly.
Chinese characters, of course, are still ideographic although not everyone realises they also have a phonetic element. Nonetheless, in days of yore Chinese writing was considered practically magical because something written in hanzi, or 漢字 in Chinese characters, could be read all over China regardless of the “dialect” of Chinese one spoke. In fact they aren’t dialects but separate languages, and beyond China the same characters were used to write Korean, Japanese and Vietnamese. In principle, they could be used to write almost any spoken language although of course they aren’t, and 國語 is particularly well-suited to being written in 漢字 because it’s tonal, isolating and purports to be monosyllabic. A language which inflected more, such as Japanese, would prove harder to write, which I presume is why that also uses hiragana, katakana, emoji and romaji. Like Japanese, Chinese has one syllable per character but also in a sense one word per character. If you’re familiar with 漢字 you’ll probably have noticed that I always use the traditional forms of the characters rather than the simplified ones, which is because I’m more familiar with them, but also because they more closely represent the meaning visually. I find simplified Chinese characters quite confounding. As I’ve said, 漢字 are partly phonetic, which makes them more suitable for 國語 than other languages as homophones in 國語 may not be in 廣東話 (Cantonese) or 客家話 (Hakka), to take two examples, because they are in reality different languages. This means that were they to be adapted, for example, for English, we would find ourselves using the same phonetic to represent the “-s”, “-en” or mutation plurals such as “teeth” and “feet” as we would to write words like “door” or “gate”, which would be less obvious than if we had our own system.
The West was quite taken by 漢字 when we first encountered them, which, surprisingly, was not from Marco Polo who never mentioned them, which is one reason why some people think his account is based on hearsay. We had had ideographic scripts in Europe, but they had been largely replaced by alphabetic ones, and in the early Mediæval period they were more phonetic for English and French, for example, than they were to become. They proved to be a major influence on the thought of John Wilkins, but not the only one.
West Afrika has a very large number of scripts, as I’ve mentioned previously. Many of them are said to have come to the promoter in a dream or vision, but it’s also claimed that this is a means of avoiding the admission that they were supposed to be secret but someone decided to reveal them to the world, needing a plausible back story. This may be true because the characters used for writing Vai, for example, also crop up in Suriname. Now these sort of stories might be interpreted in a racist kind of way: surely nothing like that would ever happen here in Merrie England? Well, not so, partly because Merrie England is in many ways politically extremely reminiscent of the kind of politics currently practiced in some parts of Afrika, and probably is to an extent even today. The idea of orishas revealing a script to select individuals in West Afrika in fact has a close parallel in Tudor England in the form of the Enochian language, alleged to have been revealed by angels to John Dee in 1583 CE and having a distinctive script, which looks like this:
It’s easy to see from this that the Enochian alphabet is closely based on the English use of the Latin alphabet at the time. However, the Enochian language itself, constructed or not, is clearly most unlike English in vocabulary and grammar. Even the claim that it was revealed by spirits but was in fact a secret script already in existence applies to Enochian because it seems to be based on an earlier script called Transitus Fluvii – “Passing through the river” – used by Heinrich Cornelius Agrippa in 1533:
Hebrew square script was the basis of this script, as can be seen from the English below the chart. There were another couple of occult scripts in Renaissance England, including Celestial and Malachim. Since at the time Hebrew was believed to be the original human language, all of them bear some resemblance to it.
Wilkins’ idea was somewhat more modern, conscious, open and practical. John Wilkins was a seventeenth century polymath and clerk and one of the founders of the Royal Society. He also proposed a decimal metric system which didn’t catch on, although it’s interesting to contemplate that if the English-speaking world had come up with it first, the US, Liberia and the “U”K might all be metric now and maybe France would be holding out with its own system instead. I should mention in passing that in Anglo-Saxon times there was also a decimal-based measuring system. Two advantages of the metric system are that it’s international and that it has a logical structure and organisation to it, and one of its features is that it tends to be used by the scientific community. All of these things were also intended to be true of Wilkins’ Real Character or Philosophical Language. His aim was to design an artificial language which was logically structured and based on concepts which would enable natural philosophers and diplomats to communicate effectively with each other, and he propounded this in detail in his 1668 tome ‘An Essay towards a Real Character and a Philosophical Language’. An example of it can be seen above, in the form of the Lord’s Prayer written in that script. It can be seen to be cursive and appears to break each word down into separate concepts, each represented by a character. It also has its similarities with Arabic script and the associated Pitman system of shorthand.
Real Character is not phonetic. It could therefore be used in principle for any language. Wilkins had found the arbitrary nature of Latin grammar an unnecessary incumbrence to communication and wished to dispose of it as inefficient. It divides concepts into forty genera further divided into differences, themselves divided into species. It can already be seen from the use of the words “genera” and “species” that the work did have some influence on future classification systems such as that of Linnæus. Wilkins also came up with a system where each category could be replaced by a syllable or sound, enabling the language to be pronounced. This made it a priori, like some other constructed languages such as Solresol, unlike a posteriori conlangs like Volapük, Esperanto and Interlingua, which build on already existing foundations, and this approach steepens the start of the learning curve.
Wilkin’s scheme didn’t catch on of course. One reason was that speaking or understanding the spoken language would be very difficult because single-letter (or phoneme) differences could completely change the meaning of a word, and the question also arises, as raised by Jorge Luis Borges, who has written on this matter at length, of whether they can even be a universal classification system of this kind. I’m not sure this last point matters too much for practical purposes, as almost any classification system organised along logical lines is better than none. We have gender in many European languages distinguishing fairly arbitrarily between types of concept, sometimes on its own as with the Danish en øre vs. et øre, meaning “ear” and a unit of currency, and these do appear to descend from an originally logical scheme involving attributes, objects and agents which later became obscured, and there are plenty of other languages with dozens of genders, where this probably helps as well, but in the case of the Universal Character the divisions would possibly be less arbitrary. That said, even for first language speakers of languages which use grammatical gender, research has shown there is a psychological influence. For instance, speakers of languages for which the word for “bridge” is feminine are more likely to see them as elegant and beautiful, whereas those for whom it’s masculine will often describe them as sturdy and protective. Were we to accept Wilkins’ version of Real Character, we might be allowing ourselves to become trapped in one person’s particular way of dividing up the world, and more creative and useful thoughts might be stifled by that.
Nonetheless Wilkins had an influence on today’s world. Roget’s Thesaurus, for example, initially composed in 1805, divides concepts up in a similar way although the schemes are probably very different. There’s also the Dewey Decimal System, used for classifying library materials, and the aforementioned Linnæan taxonomy. Even Esperanto does something rather similar to Real Character in the distinctive way it breaks words down into morphemes in order to reduce the number of words one needs to learn to become fluent.
There is, however, another aspect to Wilkins’ book which has a perhaps surprising connection. One of the sections of Part 2, chapter 4, deals with botany and breaks plants down into species, remarkably similar to how they’re classified today, which has since been verified using DNA sequencing in recent years. This contrasts with Culpepper and Gerard, both of whom have a perfectly valid classification system for them but which isn’t oriented around taxonomy so much as use and flowering and fruiting times – I won’t go into this in any depth as it impinges on one of my other blogs. However, there is another, probably earlier, work, which is somewhat reminiscent of this: the apparently not-very famous (thanks for that everyone!) Voynich Manuscript.
Again, here I run up against my demarcation problem. I said before that I think the supposèd mystery of the Voynich Manuscript isn’t really a mystery at all, but unfortunately it happens to be kind of deeply wedged in the crevasses between my blogs and there isn’t anywhere appropriate to write about it. However, I will say a couple of things about it here, partly because the last time I mentioned it, I got a fair number of hits on the post, but also because it’s relevant to Wilkins. One of the suggestions as to the nature of the writing in that manuscript, referred to as “Voynichese”, is that it is in fact a similar conceptually-based script to Real Character, although inconsistent in its application of characters to concepts and rather amateurishly done, as is the rest of the codex, which in fact is a pretty massive clue as to what it is. It seems weird how something can stare so many people in the face like this and yet not be noticed. Radiocarbon dating sets the materials whereof the MS is composed to the early sixteenth century, so it pre-dates both John Dee’s Enochian and Wilkins’ Real Character although that does make it contemporary with Transitus Fluvii, which was however not a conceptually-based script – a Begriffschrift if you like (!) (PDF alert). There’s a fair bit of botany in the MS too, of course (I mean, there would be). But this is not the interesting bit. Voynichese words are highly regular in structure. Some characters only occur at the beginning of words, some only in the middle and some only at the end of them. The same is, incidentally, true of many other scripts including Devanagari, Hebrew and Arabic, although only Arabic has all three of those features, so it isn’t a clinching argument for it being a conceptually-organised script. No words have fewer than two letters or more than ten.
If you run the above paragraph through Gender Guesser, you might get an interesting result! Or you might not.
Many people believe the MS is a hoax, but this is kind of beside the point. It may also be automatic writing, in which case it might be expected to have similar characteristics to glossolalia. I don’t know if this has been tested. Glossolalia and xenolalia are big topics in themselves which I shall reserve for another post.
To conclude, then, Real Character is a valiant and appealing effort to which John Wilkins seems to have applied his considerably honed intellect, and although as such was doomed to failure, has still left its mark on the world in numerous ways. But was he pre-empted earlier in the century by an anonymous author?
PS: Ironically, without gaming the Gender Guesser algorithm, this text only comes out at 53% male!
A Box Of Chocolates 