#hamnosys

Lauren: That we know of.

Gretchen: That we know of.

Lauren: There might’ve been a society that did a very ephemeral form of snow writing that we have lost forever. But we have records of 3 or 4 times.

Gretchen: It’s been invented a handful of times. There are a few other cases where there are scripts that haven’t been deciphered by modern humans. Maybe they’re scripts, maybe they’re not – it’s not quite clear. But it’s definitely a handful of number of times. And then once other cultures come in contact with the technology of writing, they’re like, “Oh, this is cool. Let’s adapt this to our linguistic situation,” and it gets borrowed a heck of a lot. But it only got cemented a few times.

Lauren: It’s worth saying that “3 to 4” is a bit squishy because it’s not entirely clear if cuneiform, which is a very pointy form of writing from Babylonia, somehow inspired the Egyptian system that became what we know as the hieroglyphs or if they just happened around the same time by coincidence are something we may never really fully put together. That’s a very contested situation. That’s why we can’t even pin down the number of times we think it was invented.

Gretchen: Cuneiform is the one that’s made with the sharpened reed that you push into your clay tablets or, if you’re some people on the internet, into your gingerbread because there’s some really excellent examples of cuneiform gingerbread tablets people have made, which I just wanna – yeah, it’s really great. The Egyptian hieroglyphs people have seen. But yeah, it’s unclear whether they were in contact with each other and kind of heard of each other in a very loose sense and were inspired by each other because there was some amount of contact between those two areas, or if that was elsewhere. The other two – one is in Mesoamerica, in modern-day Mexico and that area, where they had a writing system there that, again, developed into lots of different scripts as it got borrowed from different areas, of which the best deciphered is the Mayan script from the 3rd Century BCE. There’s also the Olmec script, which is probably the oldest. The Zapotec script is also really old. There’s a bunch of scripts in the modern-day Mexico area that also developed independently.

Lauren: Then the final system arose in China around the Bronze Age a couple of thousand years BCE. Because this script was mostly found in its most earliest forms on oracle bones, it’s known as the “oracle bone” script.

Gretchen: What is an oracle bone?

Lauren: They are turtle bones that are used in divination.

Gretchen: Oh.

Lauren: Yeah.

Gretchen: And, again, the Chinese script, once it developed further, it was also, yeah, influenced a bunch of the other writing systems in the area.

Lauren: I find it super fascinating, with absolutely no historical knowledge or insight to bring to this, that in these three different places that were completely separate and going about their own cultural lives writing arose at a similar time around 3,000 to 4,000 years ago.

Gretchen: Yeah! You wonder what was in the water or something. Well, and it’s partially, I think, that there’s a certain level of writing makes it easier to do things like administrative bureaucracy if you’re trying to keep track of whether people paid their taxes or – it’s a very empire-y thing to have is to develop a writing system.

Lauren: Oh yeah. And it’s absolutely worth stating that it’s not like three people in these three different locations all woke up on the same Tuesday 4,000 years ago and were like, “I’m gonna write a long letter to someone.”

Gretchen: Did they have Tuesdays 4,000 years ago?

Lauren: What you see is this emergence of, “I’m just gonna make a couple of notes so I know how much money you owe me.” Some of the earliest cuneiform tablets we have are just, like, beer supply stock takes.

Gretchen: Like, “Three oxes and this many baskets of grain” or whatever.

Lauren: I feel like it’s very human to be like, “We love writing because it’s poetry, and I can send letters to people I love,” and it’s like, no, it’s actually, “I just wanted to know how much you owe me.”

Gretchen: The king just wants to know if these people have paid their taxes.

Lauren: So, what you get is – although I’m like, “Oh, it all happened within similar millennia,” it is actually centuries of development from just keeping tabs on a few items to a fully fleshed out written system.

Gretchen: What types of things people thought were important to write down – things like legal codes and stuff like that – one of the interesting things that I came across when I was looking this up was that there’s a person named Enheduanna, who is the earliest known poet whose name has been recorded. She was the high priestess of the goddess Inanna and the moon god Nanna in the Sumerian city-state of Ur. There we go. But authorship shows up much later than some anonymous civil servant keeping track of who’s registered which grain or some anonymous priest or something keeping track of who’s made various offerings. This idea of like, “Oh, you’re gonna write poetry,” is a step later.

Lauren: Filing your tax is what is actually one of the best links you have to those ancient civilisations.

Gretchen: There’s this Egyptian named Ptahhotep – that’s “Pta,” P-T, even though I know I’m not pronouncing it that way – he was a vizier in Egypt. He’s also one of the first named writers, the first book in history – or people call him the first book in history – because he wrote these Maxims of Ptahhotep. There may have been people who were writing on more perishable materials that didn’t get recorded and stuff like that. It’s this whole process of, “Okay, I’m going to draw these little diagrams of oxen or something or draw these little diagrams of this plant or this animal or whatever to record what types of things get recorded.” But then in order for it to actually become a writing system, there’s also this step of abstraction that has to happen. This is when you start saying, “Okay, well, the word for this very easily visualisable thing” – so I’m thinking of oxen because the word for “ox” in one of the Semitic languages, I think, was something like /alef/. And so, this “ox’s head” gets transformed into, “Okay, what if this is the sound at the beginning of the word for ‘ox’s head,’” which is /alef/, and it gets transformed into our modern letter A, which is “alpha.” “Alpha” in Greek is just the name of the letter. It’s not “an ox’s head” in Greek anymore because the Greeks borrowed it form the Phoenicians. This level of abstraction that has to go from, “Okay, I’m gonna draw an ox’s head” – if you turn a capital A upside down, it kind of looks like an ox’s head.

Lauren: It’s got its little horns, which are the feet of an A.

Gretchen: Yeah, and there’re all these related languages. You know, Arabic’s got alif at the beginning, even though it doesn’t look like an ox’s head anymore. Hebrew’s got an alef, and Greek’s got an alpha, and all of these alphabets that begin with A. It’s this level of abstraction where you can use this thing to stand for this thing that was associated with an ox.

Lauren: There’re a couple of main different ways that you can relate these abstract images that you’re putting down in writing to the language that you are trying to capture. Of course, being a linguistics podcast, I was gonna bring this straight back to the structure of language.

Gretchen: Well, I think it’s interesting to look at the structure of languages in different areas of the world, and how people reflect those in the writing systems that are developed for those languages. When they borrow a writing system for a language with a very different structure, they end up doing certain adaptations to account for not just like, “Okay, languages have different sounds,” but also those sounds are organised and structured in different ways with relationship to each other. The writing systems often reflect some of that history.

Lauren: The Latin alphabet that both of us are most familiar with has a very approximate correspondence between each character of the writing system and a sound in the language. And I say “approximate” because English spelling is a wonderful historical record of how some of those sound changes have changed over time. I’m just gonna keep this upbeat. You can fall down a giant well of English writing system problems, but to get to a point where the majority of letters have a pretty stable correspondence to sounds that we recognise as phones in the language, and that allows us to write out the words of English.

Gretchen: One of the things that’s true about a lot of the Indo-European languages is that they have a particular ratio between consonants and vowels in the words, where they have a fair bit of consonants in relationship to their vowels but not a ton. You can see this in the writing system because the writing system represents consonants and vowels separately. And yet, when the Greeks were borrowing the alphabet from the Phoenicians – Phoenician is a Semitic language like modern-day Arabic and Hebrew – that alphabet only had consonants in it – letters for consonants – because the vowels were not that important. This is still true of modern-day Semitic languages is they’re often written in writing systems that don’t represent the vowels or kind of optionally represent the short vowels, or sometimes they represent the long vowels, but they’re often written in writing systems where the vowels can be omitted. That’s not really a thing you can do very well in Indo-European languages and still have things understood because the vowels carry enough information that you need to represent them somehow.

Lauren: Even when you have a phonemic script, it’s not necessary to always represent all of the sounds to convey the language.

Gretchen: Right. Then conversely, there are other languages where the vowels are even more important and, in fact, every consonant comes with a vowel or virtually every consonant comes with a vowel. In those, you often get what are called “syllabaries,” where they represent one syllable at a time, because why bother with representing each of these things separately when in every context where you have a consonant there’s gonna be a nearby vowel – or in virtually every context there’s gonna be a nearby vowel – and so you can have a symbol that just represents the whole syllable there. That’s also a structure that doesn’t work very well for Indo-European languages because they don’t have that many vowels. There’s this spot of like they have important enough vowels that you need to represent the vowels somehow but not so important are vowels that you have to represent lots of vowels all the time, whereas languages like Japanese or Hindi – well, Hindi’s Indo-European, but it’s got more vowels, I guess.

Lauren: The Devanagari writing system is inherently focused on the syllable, which is a very different approach to reading. Each character of this writing system, if there’s no vowel specified, it just comes with a bonus vowel. It’s like, “Buy this consonant, get this free letter A sound.”

Gretchen: Right. That’s partly a feature of the writing system, but it can only be a feature of the writing system because it’s already a feature of the language. A similar thing goes for a language like Chinese, where a lot of things are based around a syllable.

Lauren: Then you can go a level of abstraction further where your character in the writing system represents a word-level thing and doesn’t have a direct relationship to the sound correspondence, which is what happens with the Chinese script.

Gretchen: I think it’s important to recognise that there is a phonetic component to Chinese characters. They often make use of – especially for words that are more abstract – it’s not just like, “Oh, here’s a bunch of little pictures that we’ve drawn,” because that’s not capable of conveying abstract concepts like grammatical particles and words for things that don’t come with easy pictures. And so, making use of, “Okay, a lot of our words are one or two syllables long, so here’s a word that’s relatively easy to visualise that sounds very similar to a word that is not as easy to visualise.” We can just add a thing to be like, “It sounds like this, but it’s got a meaning more related to this,” and you can be like, “Oh, it must be this more abstract word.” The classic example, which I’m definitely gonna do the tones wrong on, is that the word for “horse” is /ma/, and the word for mother is also /ma/ with a different tone, and you can add the little horse semantic component with the woman semantic component and be like, “Oh, it’s the word that sounds like ‘horse’ but has to do with something with a woman,” and then you end up with “mother.”

Lauren: This works for languages in China because they tend to be not as long as words in English. We like to add all these extra bits of morphology within our grammar, whereas, again, you get – not a direct rule force – but you get this general tendency where the writing system kind of fits with the vibe of the grammar of the language.

Gretchen: One example of that is in Japanese where they were heavily influenced by the Chinese script, but Japanese actually does have suffixes and other little grammatical words and things you need to change about words. They made some of the Chinese characters that had formerly only had semantic things into just like, “Oh, this makes this sound, and this makes this sound,” because they needed to be able to represent that morphological information that’s not super important in Chinese but is very important in Japanese. You end up adapting a script into something else when it gets borrowed in a different context. Another interesting example here is Farsi or Persian which is an Indo-European language that’s conventionally written with the same script as Arabic except it’s also had a couple of additional letters added because Persian has a P and Arabic doesn’t. They had to create a symbol for the sound P, which is why you get “Farsi” instead of “Parsi” because Arabic doesn’t pronounce that P. So, it makes the P into an F. Sometimes you get people adding additional letters like adding a letter for P. Sometimes you get adapting whole sets of a script.

Lauren: Sometimes you lose letters. English had distinct characters for /θ/ and /ð/ until it was technologically easier to just use the characters in the printing press that English had borrowed. It’s makes me a little bit sad. But also, it makes international people – maybe it’s a little bit easier.

Gretchen: We used to have a thorn for the /ð/ sound, but those early printing presses from continental Europe didn’t have thorns on them. I mean, Icelandic still has thorns. One of the things that I think is more interesting in the closer to modern era – not strictly modern era – is cultures and peoples that are familiar with the idea of writing yet take the idea of writing and say, “We’re gonna make our own homegrown script that actually works really well for our particular language.” One of my favourites is the Cherokee syllabary, which was invented by Sequoyah, who was a Cherokee man who didn’t know how to read in English, but he’d encountered the Latin-based writing system in English. He thought it was cool that the English speakers had this, and so he locked himself in shed for several years and came up with a syllabary for Cherokee. Some of the symbols on the Cherokee syllabary look something like Latin letters, but they stand for completely different things because he wasn’t just learning to read from English. Some of them are completely different. This became hugely popular among the Cherokee in the area. There were newspapers in this in the 1800s. There was very high literacy in Cherokee country. It was really popular. It’s even still found on modern-day computer keyboards and stuff like this. You can get Windows and stuff in Cherokee. It’s this interesting example of that’s one where we can say a particular person was inspired by writing systems but also created his own thing that became very popular.

Lauren: The thing that makes Cherokee so compelling to me is not only did he come up with an incredibly elegant, well thought out, suits the language system, but that he actually got uptake as well – that the community decided to use this as the writing system that they would learn to read and write in, and that it had uptake. It’s very easy to come up with ways of improving the technology of writing but, as I think you’re fond of saying, language is very much an open-source project. You can come up with really elegant solutions, but if no one else is gonna take them up, that’s not gonna be very helpful. So, Sequoyah’s work is doubly amazing for that reason.

Gretchen: People actually made printing presses with the Cherokee symbols and were using those. Another interesting case of this disconnect between a person or people coming up with a system and actual uptake of it is Korean, which has what I think linguists generally agree is just the best writing system.

Lauren: Yeah, we’re like, “Writing as a technology is amazing. All writing systems are equally valid. But Korean is particularly great.”

Gretchen: “But Korean’s really cool.” The thing that’s cool about it from a completely biased linguist perspective is that the writing system of Korean, Hangul, the script, is not just based on individual sounds or phonemes, it’s actually at a more precise level based on the shape of the mouth and how you configure the mouth in order to make those particular sounds. There’s a lot of, okay, here are these closely related sounds – let’s say you make them all with the lips – and you just add an additional stroke to make it this other related sound that you make with the lips. Between P and B and M, which are all made with the lips, those symbols have a similar shape. It’s not an accident. It’s very systematic between that and the same thing with T and D and N. Those have a similar shape because they have this relationship. It’s very technically beautiful from an analysis of language perspective.

Lauren: I love this so much that when we were prototyping a potential script for the Aramteskan language for the Shadowscent books, when I was constructing that language, I also started constructing a script that we never used anywhere, but it was helpful to think about how the characters would write and what writing implements they would use. If you look at the script, you’ll notice that the letter P and B are very similar, but B has an additional stroke. T and D are very similar, but D has an additional stroke. Very much feature driven. And then for the vowels – it’s roughly a quadrant in the writing space – the /i/ vowel is in the top left of the quadrant, the /u/ vowel is in the top right of the quadrant, the /a/ vowel is in the bottom left of the quadrant.

Gretchen: So clever!

Lauren: It was actually just for really selfish reasons that I decided to go with a feature-based system, and that is that it was easier for me to remember if I used the features of the language and made sure that the voiced sound was always identical to the voiceless one but just with an additional stroke. It meant that I only had to remember half the characters.

Gretchen: That’s very elegant. The easy to remember bit is also true about the Hangul script because it’s got so much regularity. The famous quote about Hangul is something like “A wise man can learn it in an afternoon and a foolish man can learn it in a day.”

Lauren: So catchy!

Gretchen: There’s probably a better version of that quote. What’s interesting about it from an adoption perspective is that Hangul was invented by Sejong the Great.

Lauren: Appropriately named.

Gretchen: Who has a national holiday now because of the script. But it was created in 1443. It’s not quite clear whether it was him personally doing everything or whether he had an advisory committee of linguists, but it’s really extremely well-adapted to the linguistic situation of Korean in particular. Even though it’s just also really cool for how it represents the inside of the mouth, but it’s really well adapted for Korean. It was invented in 1443, but it wasn’t popularised in use until several centuries later because for a long time Korean was also using, like Japanese, this adapted version of the Chinese script or adapted version of the Japanese script because of the cultural influences. In the early 20th century, they were doing a much bigger literacy push in Korea to be like, “What want everyone to learn how to read.” And they said, “Okay, we’re gonna have an orthographic reform, and we’re gonna use this script which has this very nice historical pedigree but also is much easier to learn than this complicated thing that we had done that wasn’t really designed for Korean.” It’s got this historical antecedence but also it came back in the modern-day. Now, everything in Korean is written in it. It’s because it’s really easy to learn how to read and write in. The historical uptake wasn’t immediate. It wasn’t during King Sejong’s lifetime where they were like, “Oh, yeah, now we’re all gonna use his script,” people were like, “Okay, king, you’ve got this hobby,” but it wasn’t popularised until later.

Lauren: Even when there is really strong abstraction, humans have this unavoidable tendency to think about the relationship between sounds and other senses. In sound-based writing systems – Suzy Styles, who has been on the podcast before and works on perception across the senses, did an experiment alongside Nora Turoman where they looked at whether people can guess, for writing systems they’re not familiar with, which character was the /u/ sound and which character was the /i/ sound. They found that for a whole variety of scripts there is a much higher than chance – because there’s only two choices. If was completely arbitrary, it would be 50/50. But people do tend, across the evolution of sound-based writing systems, to have /u/ that has a more rounded, bigger sound has properties in the writing system that re-occur. People continue to unavoidably link the sounds of the language to the written properties of the script in a very low-level way. I’ll link to that study. It’s really great.

Gretchen: That’s interesting. It’s not gonna be 100%, but there’s this slightly better than chance relationship.

Lauren: Yeah.

Gretchen: Visual representation of physical information is also something that shows up in ways of writing signed languages.

Lauren: Yeah. Everything we’ve talked about so far, I think, we’ve talked about for spoken languages, but it is possible to write signed languages as well.

Gretchen: There are several different systems in place. Some of them are language-specific like, “Oh, this is the system for writing ASL in particular,” and some of them are kind of like your linguist, International Phonetic Alphabet trying to provide a language-agnostic way of writing signed languages for research purposes but, in a way, that’s sort of impractical, like the IPA for general use. There’s an interesting set of systems. There isn’t as much agreement among representers of signed languages in writing which amounts of information are crucial information that has to be written down and which are optional bits of information that the reader can fill in from their own knowledge of the language and the signer.

Lauren: I think it’s worth flagging that that’s not just a discussion that arises for signed languages. It’s just that those conversations got thrashed out for spoken languages four millennia ago, and we weren’t around when people were arguing about whether intonation had any role in the – or people probably were arguing because it was an emerging thing.

Gretchen: Well, when people were arguing about like, “Do we write vowels or not,” which was a big thing. Do we write vowels? Do we write intonation? And punctuation followed quite a bit after – you know, punctuation wasn’t as much of a thing for several of the early centuries and millennia of writing. They didn’t do punctuation. There’s some level of ongoingness that’s still there. If you think about the internet efforts to try to write tone of voice very precisely and communicate sarcasm and irony and rhetorical questions very precisely, there’s some level of ongoing debate that’s still happening in the spoken language context but not nearly as much as is still happening in the signed language context.

Lauren: Also, just because of the way that signed language communities tend to be embedded within larger spoken language communities, people who sign as a primary language tend to also be educated in the mainstream spoken language, and so literacy gets developed in, say, a language like English.

Gretchen: I think that’s the case for a lot of smaller spoken languages as well where sometimes there’s this imperative of, “Okay, we want to be able to write things to each other” or something, but if there hasn’t been a history of a lot of published literature in that language that you’re trying to read, then it becomes a question of, “Should we teach this in school,” because there isn’t literature there, even though there would be oral literature. It becomes a chicken and egg problem of which comes first, or which do you start teaching first, when you’re constantly comparing stuff against a few very large spoken languages that have this very long writing tradition. It shows up in languages with a newer writing tradition.

Lauren: Education systems have a massive influence there. My grandmother, actually her strongest written language is German. Even though she and her sister speak to each other in Polish, they would write to each other in German because that’s the language they had been educated to write in. Even with people who don’t speak minority languages, the influence of the education system there is so massive.

Gretchen: Reading and writing, they’re separate skills even though they’re often taught together. Sometimes you can read a language that you can’t write or something like that. But it’s a big question. With signed languages, because video technology is now available, if we’d had good audio recording technology 4,000 years ago, the pressure to develop writing systems for spoken languages might not have been as strong – probably wouldn’t have been as strong – even though there are other useful things that writing can do even in the audio-video era. It’s easier to be like, “Well, you can just make a video of the signer,” and then you’d know exactly what they were trying to say and exactly how they wanted to say it. You wouldn’t have this level of abstraction of are you gonna try to write it down in a way that imperfectly represents what a person is gonna do when they’re producing it. It is still interesting looking at some of the signed language writing systems. Some of them, like Stokoe notation and HamNoSys, which stands for “Hamburg Notation System,” they try to very physically represent the characteristics of the signer – where their hands are, where their face is, and things like that. There’s another one that I can’t find the name of that is based on the ASCII alphabet, so you can type it into search engine boxes, which has some advantages as well but represents things more abstractly. It’s got this link with Korean, which was representing this very physical aspect of what the mouth is doing. Several of the signed language writing systems like Stokoe and HamNoSys also have this very physical representation what the body’s doing when it’s being produced. But I think they’re more popular among researchers than they are among actual D/deaf users who tend to use video a lot.

Lauren: I encounter Stokoe and HamNoSys in the gesture and signed linguistics literature. I haven’t really seen them too much outside of that.

Gretchen: I think that it’s easy to conflate a language with its writing system because we’re so used to thinking of English as sort of inextricably linked to the Latin alphabet. But there isn’t a reason, in theory, why you couldn’t write English in the Greek alphabet or in the Arabic alphabet or in a very adapted version of Chinese characters where you’d have to do a lot of adaptation. The same thing is true when you write languages that don’t originally use the Latin alphabet and you have romanisations of them. Writing systems are just as much political and contextual. Some of them have this very tight structural relationship to the properties of the languages they represent and some of them have looser relationships because they’ve been adapted to it later.

Lauren: It’s this slightly looser relationship to language as it’s spoken or signed that means that linguists don’t always include writing systems in, say, an Introduction to Linguistics course. We don’t often talk about writing systems. But when we were putting together the Crash Course series, we ended up making writing the topic of our final episode for the series.

Gretchen: I think partly because people are really interested in it, so why not do something about writing, and also because I think that you can use writing systems as a window into some of the interesting structural features of different languages and how the writing systems represent that. As somebody who’s really interested in internet linguistics and the rise of informal writing and how we represent tone of voice and things like that in modern-day writing, and that’s still a moving target evolutionarily speaking, I think it’s interesting to give that linguistic lens on writing systems even though they are imperfect representations of the languages that they represent.

Lauren: “Writing Systems” is Video 16 of Crash Course linguistics, which is wrapping up this month. If you’ve been holding out to watch all 16 of those episodes, you’ll be able to do so very soon or perhaps even now thanks to the temporal vagueness of podcasts.

Gretchen: Crash Course is the YouTube series that we’ve been working on basically all of 2020. It’s especially popular with high school or undergraduate teaching. If you know people that age, or who teach people that age, that may be a useful thing to send to people. We hope that people find it useful as a resource for self-teaching or for instructing in various capacities.

[Music]

Lauren: For more Lingthusiasm and links to all the things mentioned in this episode, go to lingthusiasm.com. You can listen to us on Apple Podcasts, Google Podcasts, Spotify, SoundCloud, YouTube, or wherever else you get your podcasts. You can follow @Lingthusiasm on Twitter, Facebook, Instagram, and Tumblr. You can get IPA scarves, “Not judging your grammar, just analysing it” mugs, and other Lingthusiasm merch at lingthusiasm.com/merch. I tweet and blog as Superlinguo.

Gretchen: I can be found as @GretchenAMcC on Twitter, my blog is AllThingsLinguistic.com, and book about internet language is called Because Internet. Have you listened to all the Lingthusiasm episodes and you wish there were more? You can access to 48 bonus episodes to listen to right now at patreon.com/lingthusiasm or follow the links from our website. Patrons also get access to our Discord chat room to talk with other linguistics fans – like, do you remember learning how to read – and other rewards as well as helping keep the show ad-free. Recent bonus topics include an AMA with a lexicographer and our favourite stories and anecdotes that we just didn’t have time for in some of the earlier episodes. Can’t afford to pledge? That’s okay, too. We also really appreciate it if you could recommend Lingthusiasm to anyone who needs a little more linguistics in their life. And, hey, tell your other favourite podcasts that they could a linguistics episode, and get us on! It’d be fun.

Lauren: Lingthusiasm is created and produced by Gretchen McCulloch and Lauren Gawne. Our Senior Producer is Claire Gawne, our Editorial Producer is Sarah Dopierala, and our music is “Ancient City” by The Triangles.

Gretchen: Stay lingthusiastic!

[Music]