Even if writing isn't something that we were built to do, and isn't part of our core language system, there's no question that it's an important part of our linguistic and cultural lives, and worthy of study. Proto-writing systems, like the Vinča symbols, are 6,500 to 8,000 years old, but the earliest full-fledged writing systems date back around 5,000 years. These earliest systems were generally concerned with keeping track of accounting and record-keeping, but before long, the potential of writing to move beyond just simple dry accounts became obvious, and a number of systems developed.
Writing comes in a number of varieties. Alphabets are systems where, ideally, every consonant and vowel in your language is represented by a single character. This goes sound by sound, as with the Latin or Greek alphabets. Another option is to have a system represent every consonant and vowel pair in your language, in which case you have a syllabary, as with the Japanese kana systems or Inuktitut.
Another common solution is to focus on the consonants in a given word, and leave the vowels as a secondary concern. People have developed two kinds of these systems. The first is the abugida, where the characters are consonants whose templates are changed in regular ways to indicate what vowel is to be said. Abugidas are used in Ethiopic languages like Amharic and Tigrinya, as well as in scripts of the Indian subcontinent, like the Devanagari script used in Hindi. The other is the abjad, which are scripts that don't really use any vowels at all! While there are extra marks available in these systems to show vowels, texts for adults don't use them, and readers are expected to work the words out on their own from context. Languages that use abjads include Hebrew and Arabic.
Finally, you could use a logographic system, where instead of focusing on pronunciations, you focus on meanings: each character represents one morpheme. Chinese scripts, which are used for a number of Sino-Tibetan languages, as well as in Japanese and Korean, are logographic, as were hieroglyphics, for the most part. Hieroglyphics also had a phonetic component, as well, which meant you could read characters for their meanings or for their pronunciations.
Like spoken languages, written languages have changed over time. For example, the Greek alphabet have evolved into other systems, like the Latin and Cyrillic alphabets. Also, the Chinese character set has been simplified in different ways for mainland Chinese languages now, as well as for Japanese. The simplifications put into place for the two languages have been different, too. No part of language stays fixed forever.
We talked in the episode about how Chinese characters have been adapted for Japanese in terms of the logographic kanji system. But that's not the only way that kanji have changed over time to fit the Japanese language. The Chinese languages and Japanese are quite different in terms of their sounds and the shape of their words. Chinese languages are tonal, meaning that the pitch on any given syllable matters for its meaning. We can find minimal pairs, or words with different meanings that change only one thing about their pronunciation, that differ only in their tones. For example, in Mandarin, 麻 [má], with a tone rising from middle to high, means "hemp", while 骂 [mà], with a tone falling from high to low, means "scold". There's nothing different about the consonant or the vowel for how you articulate them, just in terms of your pitch while you're saying them.
Japanese, on the other hand, has a pitch accent system. Pitch accent is like if you took tonal changes, and smeared them across a number of syllables over the course of the word. So rather than the tone happening on the one syllable itself, the change happens over the course of the word. Just compare [saꜜke] "salmon", versus [saꜛke] "alcohol, sake". For the fish, the first syllable is higher than the second; for the drink, the second syllable is higher than the first. Native Japanese words with pitch accent are thus generally longer than Chinese tonal ones, but that in and of itself doesn't mean that you couldn't just move the Chinese characters for use in Japanese.
The bigger problem is that Chinese and Japanese morphology also differ. Chinese doesn't have much in the way of bound morphology; the morphemes that the language uses for functional purposes, like negation and tense, are free to show up in different places in the sentence. Japanese, on the other hand, has a lot of affixes that have to be attached to the roots for them to both the root and the affix to be used. This includes all the verbal morphology, like tense and negation, as well as the case markers that show what roles nouns fill in a sentence. The verb root for dance may be [odoɾ-], but you can't ever say that! You have options, like an [ɯ] to make [odoɾɯ], to dance or dances, or an [anai] to make [odoɾanai] don't dance, but you have to attach something. And the [ga] or [o] or [ni] that you use to show whether a noun plays the part of a subject, object, or indirect object in Japanese don't have analogues in Chinese, but are very important for interpreting Japanese sentences.
So when Japanese first adopted Chinese script, along with a lot of other Chinese culture, it wasn't immediately obvious how best to put these together. Matching up the meanings of the Chinese logographic characters to Japanese was one thing, and not that big of a problem. But how do you deal with representing these grammatical bits, when the language you're borrowing your writing system from doesn't have anything like those sorts of morphemes, and particularly not bound morphemes?
The first solution that was used is known as Man'yogana. This system was used from at least around 470 CE for a few centuries, and can be found in very early Japanese texts, like the Man'yoshu, an early Japanese poetry collection, and the Kojiki and Nihon Shoki, the two earliest chronicles of Japanese religion and mythology. In this system, the Chinese characters were used more for their phonetic properties than their meanings. So a set of characters like "多太古要久礼婆", from Man'yoshu poem 17/4025, was read as [tada koekureba], and meaning "just to pass over," but none of those characters have anything to do with that concept. 多 means many, 太 means fat, 古 means old, etc. It's not that you couldn't have characters be close to their meanings from Chinese, but the point was to sound more like the Japanese forms.
But this was never really standardized, and many different characters could be used to express the same Japanese syllables, so the Man'yogana wasn't really the best system. While the use of kanji to express meanings never went away, using Man'yogana to show how words should be pronounced wasn't very practical. So instead, over time, two different syllabaries that matched Japanese phonology better developed, by treating the kanji in different ways.
One, katakana, was created by monks who wanted a solid pronunciation guide. To that end, they broke off pieces of various kanji that had the sounds they wanted. So for katakana, why not just take the top part of the kanji being used for [ta], 多, and just write タ? Or the left part of the one being used for [ne], 祢, to get ネ? These have the benefit of being different from the kanji, but also making reference back to the original form. And with these examples, you know what you should be pronouncing. Finding the shape of the katakana in the kanji's not always that straightforward, though: [ki] キ was derived from the kanji 機.
The other syllabary, hiragana, was derived from more casual cursive "grasshand" writing, and became used more widely among women, who had less access to the full body of Chinese learning and education (although it didn't stop them from doing some beautiful writing; The Tale of Genji, arguably the world's first novel, was written in a combination of hiragana and kanji). They similarly used kanji that were associated with particular pronunciations, but derived them from the cursive forms: the hiragana い [i] was derived from 以, and the character や [ya] from 也.
Because the two syllabaries were derived in two different ways, characters in both systems could come from the same kanji, and end up looking quite different. For example, the character for [me] in katakana, メ, and the one in hiragana, め, both came from the kanji 女. Katakana took the lower right section, and hiragana was more the character as a whole. The same can be seen for [ɯ]: while both characters derive from 宇, in katakana, it's ウ, from the crown part; in hiragana, it's う, from the whole character in cursive. For a complete look at this, try looking at this site, which shows the progression of the characters well.
So you can adopt characters of one type, logographs, like Chinese characters; import them for their meanings into Japanese kanji; use them for their pronunciations, like in Man'yogana; and then break or modify the character systems, like in the two Japanese syllabaries, katakana and hiragana. Just like the rest of language, writing can evolve to meet the needs of the people using it.
So how about it? What do you all think? Let us know below, and we’ll be happy to talk with you about how writing systems have developed and changed over time. There’s a lot of interesting stuff to say, and we want to hear what interests you!