If I asked you to tell me structure of a haiku poem right now, you would probably say that it is three lines consisting of five, seven, and five syllables in that order.

 If you were talking about the traditional Japanese method of writing haiku’s, that wouldn’t quite be right. To understand how haiku’s are written, we need to talk about Japanese syllable structure.

Japanese is a language that prefers simple CV syllables. This means that a typical syllable in Japanese consists of a single consonant sound (C) followed by a single vowel sound (V). All of this is not to say that there are no exceptions, because it just wouldn’t be a rule if there weren’t exceptions. The moral here is that the Japanese language does not like to have consonants at the end of a syllable (known as coda consonants).

To see this in action, lets look at some of the English words that have made their way into Japanese as “loan words”. Words like “glass” and “restaurant” in English become ガラス “garasu” and レストラン “resutoran” when they are adapted into Japanese.

You will notice in a word like “ga.ra.su”, the entirety of the word conforms to this strict CV structure, while in “re.su.to.ran”, the majority of the word is CV with the last syllable having a consonant at the end. Like I said, there are always exceptions to rules. It has to do with the type of sound that determines whether it can be a coda consonant or not.

In Japanese, coda consonants are permitted for nasal sounds like /n/ and /m/, but it is only the case that they are permitted to be coda’s. It is not the case that they will always be coda consonants if they show up after a vowel. There is a whole complicated set of circumstances that determine the actual structure with them, but for now, lets just say that nasal consonants are special in Japanese.

So now that we know a little more about Japanese syllables, let’s talk about what a syllable actually is. You can think about syllables as more than just consonants and vowels, and start to think about them like a hierarchical structure like this diagram below.

In a structure like this, syllables are composed of several interlocking parts. At the centre of every syllable is a vowel. Borrowing a term from science class, we refer to the vowel of a syllable as its nucleus. This is why children are taught to count the vowel sounds in words to determine how many syllables there are because every syllable requires a vowel (with exceptions as always). Everything before the vowel in a syllable is known as the onset, and everything after the vowel within a syllable is known as a coda. Taking the nucleus and the coda of a syllable together, we get a unit that is known as the rhyme of the syllable, which makes sense if you consider the fact that this is the part of a syllable that makes it rhyme with other syllables.

In a language like English, syllables can be composed of complex onsets and complex codas. You can see this by looking at a word like “strength” which is a single syllable word that has a lot of consonant sounds on either side of the vowel. With the Japanese tendency to prefer CV syllables, you can start to see a pattern where syllables are only allowed to have simple onsets composed of a single sound, and only in rare cases are they allowed to have coda consonants.

If you recall from my post on expletive infixation {HYPERLINK}, I talked about the concept of prosodic feet and how they are composed of at least two syllables, with one of those syllables usually being stressed. Well, there is a unit of measurement that we care about that comes between the syllable and the foot. This unit is known as a mora, and it is represented by the Greek letter ‘mu’ (μ).

Mora’s are what give us the concept of “syllable weight” which boils down to the principle of syllables with more morae are “heavier” than syllables with fewer morae. When it comes to assigning these morae to a syllable, they can only be associated with the rhyme of the syllable. No matter how complex the onset of a syllable is, it isn’t going to contribute any morae to the syllable as a whole. Coda consonants on the other hand can contribute weight to the syllable. In Japanese, where coda consonants are already so rare, those coda consonants almost always contribute to the weight of a syllable.

Morae are usually assigned on a one-to-one basis where a vowel will have a single mora, and every consonant sound in the coda will be assigned one mora. If you have a long vowel though (a vowel with added stress to it that makes it sound “longer” or more prominent for a longer duration) those will be assigned two mora’s.

To hear what this sounds like, listen to these two clips of the Australian pronunciation of the words “cut” and “cart”.

Notice how both of these use the same vowel sound, but the vowel in “cart” is longer than the vowel in the word “cut”. This is an example of contrastive vowel length where two words are pronounced using the same vowel, but there is a difference of length that gives rise to the distinction between the two words.

Traditionally, Japanese has syllables with no codas a single mora on the vowel. However, Japanese also has long vowels meaning that they can have bimoraic syllables without a coda consonant. Looking back at “garasu” and “resutoran”, you can count it out and see that “ga.ra.su” has three syllables with a single mora each, while “re.su.to.ran” has three syllables with a single mora each and one syllable that has two mora’s because it has a coda consonant.

Getting back to haiku’s (that was what we were talking about right?), a traditional Japanese haiku is not composed of five syllables followed by seven syllables followed by another five syllables. Many scholars argue that these haiku’s care about the number of morae in the words rather than the syllable count. So if you had a total of four syllables in the first line of a haiku, but one of those syllables contained a long vowel or a coda consonant, this would be a perfectly acceptable haiku line.

Now this is not to say that you should email your high school English teacher and tell them that they lied to you for all those years, but I hope this was at least a little informative and interesting to you. There are so many interesting things that I only just barely scratched the surface on here. Japanese is an incredibly interesting language to study, and the phonology that I talk about here is skipping over some very cool stuff. If you have the time, I suggest you read a little more about it here:

All that said, come back for more next week and in the future as I am sure that we will be returning to Japanese phonology again at some point. If you have any topics that you want to know more about, please reach out and I will do my best to write about them. In the meantime, remember to speak up and give linguists more data.