I wrote earlier about the problems of deciphering Akkadian, going from transliteration to normalization to translation. And you run into similar problems with Sumerian. Of course, if you want to capture the sound of Mandarin, you are left with pin yin, which fails to distinguish homophones that are clearly marked in the hanzi. Finally, there’s the perennial problem of bough, cough and enough. But even in a language with regular phonetic spelling like Spanish, problems arise. And with good reason. How we make language and how we interpret it differs.

In his Manual of Sumerian, Hayes talks about the difference between phonetic and morphological transcription. That is, when turning the Sumerian characters into a Latinized notation for the sake of organizing your thoughts on what represents what, is it better to capture the morphemes – the units of meaning – or the phonemes – the units of sound? Is a compromise possible? And if so, what?

To turn the question into English, let’s take a phrase:

I am going to go to the zoo tomorrow.

I’m gonna go to the zoo tomorrow.

Aym guhnuh gohtuhthuh zoo t’morrah.

The first of these is a morphological transcription: It captures the elements of meaning. The last is a phonetic transcription of what was said. The middle is a compromise: It clearly breaks out most of the morphemes and the one place where it runs things together is sort of an official misspelling. Which is best? Unless you like to read aloud, the first or second is far more convenient. It means that you can see what units are being put together and it assures that people with different accents will write the same thing. This makes writing a trusty way of communicating meaning, but not a trusty way of communicating speech.

There is something useful to be gleaned here: What makes the Cuneiform languages so hard to decipher is that there’s not a good balance between their representation of morphemes and phonemes. But though they are an extreme, they point to a problem even with a language like Spanish: To truly understand the spoken tongue and produce it requires a sort of normalization where sounds run together in order to figure out where one word ends and the next begins and just what the endings and beginnings of those words are. Getting at home with your language, then, is going to require going beyond text and trying to get a sense of how your language sounds in your head and coming off your tongue if you truly want a feel for the language you’re learning.