Jochen Gos emailed me awhile back with news of her new book Icon-Typing and website Icon-language which is a font that creates pictograms when a person types in their own language’s sentences. Unfortunately, it’s taken me a rather long time to finish this post about it.
While it’s a fun idea, and reminiscent of other projects of the same nature, I should say that I do not consider any of these approaches to be “real” visual languages. And, perhaps dissapointing to some, I don’t consider any enterprise that strives for a “universal pictographic writing system” to be a fruitful endeavor for multiple reasons.
I do think that the idea of a pictographic language stems from a deeply felt intuition that there is indeed a visual modality language, but cannot reconcile it with the assumptions our culture makes about graphic signs and language. Among the problems holding back these types of endeavors:
1. Semiotics – There is a general misunderstanding, both in and outside of academia, that there are only two types of graphic signs: those that map to “ideas” – mistakenly like Chinese or pictures – (ideographs) and those that map to sounds – like alphabets (phonographs). This is a gross over generalization that I attacked in my thesis ¡Eye græfIk Semiosis!, trying to show that there is a much more graded system of expression at work here.
2. Grammar – All “univeral” pictographic languages must piggyback on the grammar of existing verbal languages. Verbal languages are vastly different in their grammars so much so that no possible universality could ever be reached. There is a deeper reason for this though: dividing pictograms into linearized individuated units denies the grammatical qualities inherent to the actual visual form. A grammar must arise up out of the signs themselves, not be imposed on signs from the outside (see again my paper ¡Eye græfIk Semiosis!). An Example: A pictographic system might use a marker to show plurality paired with an icon. So, a pictographic symbol might express “men” like [MAN PICTURE]+[PLURAL MARKER] (not necessarily linearly)as opposed to just showing multiple men couched in an environmental setting. In one, the concept arises directly from the sign, in the other it’s latching onto the way that verbal languages do it.
3. Morphology – Believing that pictograms from all languages can easily substitute in for the words of a language presupposes that all languages have “word” units like English. What would this do for a polysynthetic language like Greenlandic or Dene that chunks multiple meaningful units into a whole syntactic unit (i.e. a single “verb” in these languages might have the same quantity of information as an English sentence). Chunking up visual pieces like this betrays a bias for how European languages break up units, not all languages.
4. Linguistic relativity – To believe a universal writing system is possible assumes that all languages structure concepts the same way. This is simply not true. The aforementioned Dene has several forms of the same verb meaning, each one reflecting a different type of object (for instance, the verb “give” changes based on the texture and character of what is being given – whether its animate, granular, squishy, flat, etc). Less “exotic” differences are like that Japanese lacks plurals and definitive articles. Plus, given how widely languages differ, which morphemes would be worthy of visual conversion? Do you convert all of them (including things like transitivity or definiteness?) or just those most “visually relevant”?… and how would that not be a subjective decision based on the preferences of the speaker of a particular language? A speaker of English might be more inclined to need particular features (like plural) that differ from a Japanese person (like politeness). A universal writing system assumes that concepts are universal to express and do so in similar ways: a big assumption.
5. Transparency and Iconic Bias – There is a general belief that Iconic signs that look like what they represent are wholly transparent in their meaning and non-culturally relative in and of themselves. This is also untrue. While most all cultures seem to be able to decode what “realistic” drawings represent, they often are subject to entrenched variability. For instance, Australian drawing systems usually feature aerial viewpoints, which can lead to odd interpretations of lateral viewpoint representations. Furthermore, simplfying the “concept bundle” of images down to a one-to-one concept-to-sign ratio is extremely difficult.
Finally, the biggest issue here is that the visual form is implicitly considered as lesser than the verbal. It uses the verbal form to be translated into the visual without stopping to think that the visual form might already have its own visual language system naturally that doesn’t need a verbal connection at all. Writing systems themselves are not a bad thing, but as an importation of one expressive modality into another they will always be subject to the constraints of the imported system.
A real, natural Visual Language should be able to stand alone without needing such connections to lean on, and would reflect the diversity of different cultures as well.