Timing, AKA isochrony, is a key part of producing (and perceiving) non-written language. In English, we time our speech by the stress on the words, usually smushing together the unstressed bits so that there is equal delay between each stress. English isn’t the only language to do this– German, European Portuguese, Arabic, and many more do, and for this reason are called “stress-timed” languages. Timing can also be divided by syllable, where each syllable gets an equal amount of time. These “syllable-timed” languages include Spanish, Brazilian Portuguese, Korean, and more and are easily identified by their bouncy, consistent sound.
You may have noticed two things so far from this. First, that Portuguese can be both stress and syllable-timed depending on the dialect, and second that Spanish is considered syllable-timed when some dialects/accents break this rule. I show this to you to demonstrate that languages are not necessarily rigidly one or the other, that dialects and accents can alter this, and that some are actually quite strict in comparison.
But now, I present to you a less intuitive timing: mora-timed languages. Japanese and Hawaiian are the most commonly mentioned examples of this type of timing, but some African and Oceanic languages also use this. But what is a mora, you ask? Great question, and it is going to take a little explanation of linguistics to get there. In short, morae are ways of breaking down a word. Just as you can break a word down by syllable and phoneme, you can also break them down with morae.
Essentially, there are three possible parts of a syllable: the onset, nucleus, and coda. The nucleus, as the name suggests, is the core of the syllable and majority of the time is a vowel. The onset and coda come before and after the nucleus, respectively, and can be consonants or consonant clusters. Together, the nucleus and coda form something called a “rime” or “rhyme,” which makes sense given the colloquial meaning of the word.* For the word “cat,” the letters “a” and “t” make up the nucleus and coda, and therefore the rhyme, and are what we typically use to rhyme words together. A cat plays with a bat while wearing a hat. The “at” is what we care about here. I should also mention that syllables don’t need an onset or coda; they just need a nucleus to count.
Now we are going to skip to the unit smaller than a mora, which is a phomeme. Phonemes are the sounds that are made. They can be consonants (C) and vowels (V). If we used the word “cat” again, it would be written as CVC. Mind you, when you see a diphthong, or two vowels together that make a new, combined sound, these are not usually VV, but are written as V since only one sound is made. With consonants, the same can apply. Digraphs are when the written language uses two characters to represent a single sound (graphemes are the ways we write phonemes, hence the word “digraph”). For the word “aisle,” you would write VC, as there are only two sounds being heard. The “ai” in that word is both a diphthong and a digraph. Yet for the word “taste,” you would write “CVCC” as there is one syllable, yet four sounds are heard. A slash is used to separate syllables in multi-syllabic words.
Now, the moment we’ve all been waiting for: the mora. Morae sit between syllables and phonemes, sometimes scooting on their chair a little closer to one or the other depending on the word. They are more complex than just a single sound, but also are more specific than a syllable in terms of timing, specifically in regards to their vowels. In Japanese, for instance, double vowels don’t typically get combined as a diphthong but instead are given double the length of time when spoken. This is not typically the case in stress or syllable timed languages, where such vowels would be pushed into their respective stress or syllable patterns. For instance, take the word Ōsaka, as in the city name. This word has three syllables, yet has four morae. Since the first “o” is doubled (written as “ō” here or “おお” in hiragana), it is given two morae worth of time when spoken (four morae total for the word). Another example is the word “konkyo,” which means “evidence” in Japanese. This word has two syllables, kon-kyo, but has three morae, ko-n-kyo. The “n” is its own mora since CVC would break Japanese’s rules. To show another instance of what can and can’t work, let’s look at the word “pineapple.” The English word is split into two syllables as CVC/VCC. In Japanese it is written as “painappuru,” is 6 morae long, and is CV/V/CV/C/CV/CV (pa-i-na-p-pu-ru). This is because Japanese does not allow true codas in their language. Again, only V and CV are allowed, with some exceptions regarding nasal and geminated sounds. Often times the number of kana in a word correlates with this.
The tricky part is that some people say that Japanese (and other mora-timed languages) have syllables as well as morae, while some argue they forfeit their syllables. From what I understand, they do indeed have both, but morae are used with a native mindset of the language while syllables are for more of a linguistic breakdown. AKA, Japanese speakers think of word breakup in terms of morae, hence why haiku poems are written in 5-7-5 morae, while syllables are used by linguists who want to breakdown how the word sounds and functions. Note that other languages that are somewhat mora-timed may use and think of both semi-equally, though. I might be wrong on this, but my sources are telling me this is what it seems to be. For instance, take the word “konkyo,” which means “evidence” in Japanese. This word has two syllables, kon-kyo, but has three morae, ko-n-kyo. The “n” is its own mora since CVC would break Japanese’s rules.
Overall, mora-timed languages have a quite distinguishable sound, and one big part of that is thanks to their unique isochrony. Morae are a hard concept to comprehend, and if I am being honest, one I am still trying to fully grasp, but hopefully this provides somewhat of an explanation. I may return with edits if I figure out something new.
Happy reading,
-Beppa
Some sources:
-Starr, R. L., & Shih, S. S. (2017). The syllable as a prosodic unit in Japanese lexical strata: Evidence from text-setting. Glossa: A Journal of General Linguistics, 2(1), 93. https://doi.org/10.5334/gjgl.355
-This presentation, that I believe is part of the paper above: https://slideplayer.com/slide/4455732/
-http://blog.livedoor.jp/chikocafe-langcafe_japanese/archives/1016124277.html
-https://en.wikipedia.org/wiki/Mora_(linguistics)
*Disclaimer: the term “rhyme” in this way may only apply to English or a select number of languages, as I found this info while researching English phonetics. There may be other words or breakdowns out there for other languages.