Building a Fantasy Language 2: Sound System - Stress and Friends

In my last post, I talked about different factors to consider when you create the writing system for your fantasy language.  Now I’m going to talk about some things to think about when you’re deciding how your language should sound.

In my experience, usually the first thing I want to do when I begin working on a language is to start a word list.  I’ve hit a point in my story where someone should be speaking language X… usually saying something like “Hello, how are you?  The weather is good today.”  But if I just start making up random words and throwing them onto the page, they may not seem like they’re all from the same language.  All too often, they look like just what they are: random words thrown onto a page that in no way belong to one coherent language.  And we’ve all read published books where, yup, the author just threw random sounds onto the page and called it language.  Painful!  (Editors, please don’t let us do this… )

So what’s the critical factor that makes your language convincing to the reader?  For a fantasy language, the critical factor is probably sound.  Not grammar, but sound.  (The grammar geek in me hates to admit this.)  The reader who will actually go to the appendix and read through my explanation of verb conjugations in Allahoallurian is a rare bird.  The readers who will recognize at a glance that I put zero effort into my language’s sound system are far more common. 

Thus this post.  Or, actually, posts—I decided to split the current topic into three parts, to keep this from being so long that no sane human would want to read it at a sitting.  The three parts are as follows: first, the topic of prosody.  This is a ten-dollar word that just means things like stress, pitch, tone, on a word level or on a phrase or sentence or utterance level.  (I’ll just talk about the word level, which should be all that you need for a fantasy language.)  Second, the topic of phonology, which deals with what sounds you have in a language, how they are organized, and how they interact with each other.  Third, the topic of syllable structure.  How can sounds be put together into words?  (One caveat about my breaking the topic up like this.  Different issues in the linguistics of a sound system are all tangled together.  Complex stress rules are sensitive to syllable structure, syllable structure is sensitive to phonology.  So you may eventually want to glance over all three sound system posts, even if you originally thought you only needed to look at one.  If there’s something not explained in one post, maybe it’s explained in one of the others.  If not, leave a comment!  :)

Today… prosody.

Prosody (stress, tone, intonation, pitch, accent, etc.  etc. etc. etc.)

Prosody is a very complicated topic that dozens of people have spent their entire lives studying.  Fortunately for me (since prosody always confused me), only a few aspects of prosody are important for building a fantasy language.

The most important thing that you need to know is that there are three prosodic types of languages (as far as I know).

1. Stress languages like English
2. Tone languages like Mandarin Chinese and many African languages
3. Pitch-accent languages like Japanese

In the three sections below, if you get tired of the technical stuff you can skip to the end and see my quick advice on how to handle each one of these for a fantasy language.

Stress languages

What does it mean to stress a syllable?  In various languages, people emphasize a syllable by making it louder, longer, higher pitch, lower pitch… okay, basically people do all sorts of different things in different languages.  But somehow or other they make that syllable stand out from other, unstressed syllables.

We are pretty used to English word-stress rules.  English usually assigns stress (decides to stress a syllable) based on things like how many syllables there are in the word, what structure they have—that kind of thing.  (ENglish Usually asSIGNS STRESS…) Occasionally, you find words that have a built-in stress pattern that doesn’t seem to be based on the rules.  For example:

            Nouns                                    Verbs

COMpact                               comPACT

PERvert                                 perVERT

CONtract                                conTRACT

Obviously these words aren’t being assigned stress based on length or syllable structure, since they exist in contrasting pairs!  Words drawn from foreign languages may also mess with the stress rules by keeping the stress they had in the language they were borrowed from.

Normally, however, words in stress languages follow the rules, which assign stress based on number of syllables and syllable structure.

STRESS        AU-to              CA-rousel

TRUE             EN-glish        BEAU-tiful

RUN               JUMP-ing      PAR-rying

GREEN          BLIND-ly        AR-dently

IN                    O-ver             

There will be a primary stress and, once the word is at least three syllables long, there may be a secondary stress too, also assigned according to the rules.

Different languages have different rules, but three common places to stress a word are the last syllable, the next-to-last syllable, and the first syllable.

Even with rules, there are complications…

Different types of words in a given language may have different sets of stress rules.  Verbs may have different stress rules than nouns—in fact, they usually do.

Some types of prefixes and suffixes, called clitics, do not take stress and do not participate in the stress rules.  For example, even if a language has a strong stress rule that states “stress the final syllable,” a suffixed clitic will not get stressed.  Instead, the last syllable before the clitic will get stressed.

Fantasy Language Stress

If your language is a stress-language, I would recommend that you choose which syllable you want to stress and have all your words stress that syllable.  If you want to make your language a little more complex, choose one syllable (such as the next-to-last) for verbs and another (such as the last syllable) for all other words.  You can mark a stressed syllable with a little acute accent: ˊ.

ye-ru-sha-laím                      mam-la-káh               maʕ-yán

If you make up a word and try to stress it according to your rule, but have a strong feeling that the stress “just isn’t working” for this word, the syllable structure of the word is probably causing the problem.  (I’ll be posting about syllable structure shortly.)  For now, just keep in mind that heavier syllables (ones with long vowels, diphthongs, or endings involving more than one consonant) tend to attract stress.

Tone languages

In tone languages, speakers use changing pitches (like changing notes when you sing) to mark different syllables.  Instead of the tone being assigned based on the length or syllable structure of the word, the tone is assigned because it is integral to the meaning of the word.  In Mandarin Chinese, ma with tone 1 is word that means something completely different from ma with tone 2, and both mean something completely different from ma with tone 3 and ma with neutral tone.

How many different tones can you have in a tone language?  In my experience, languages can have from two to eight tones, with four being the average.  Imagine having to distinguish between eight different intonations of ma!  Generally speaking, languages that allow longer words and more complicated syllable structures will have fewer tones.  Languages with shorter words and simpler syllable structures will have more tones.

Many African languages have two tones, a high tone (H) and a low tone (L).  They need only two pitches, a high pitch and a low pitch. 

But in other African languages, even though they still have only a high pitch and a low pitch, they have four tones!  How does this work?  Well, they have

tone HH – stays at a high pitch

tone HL – starts at a high pitch and goes down

tone LH – starts low and goes up

tone LL – stays at a low pitch

Some languages add a third pitch, that is between the high and low pitches—the mid pitch (M).  Now how many tones can we have?

HH, HL, LH, LL +

MM – stays at the mid pitch

ML – starts mid and goes down

MH – starts mid and goes up

LM – starts low and rises to mid

HM – starts high and drops to low

With three pitches, we can theoretically have nine tones!  In reality, languages with H, L, and M pitches usually will not have all nine of these tones.  Some of them sound really similar! 

Are there any more possible tones?  Actually, yes!

LHL – starts low, rises, then drops again

HLH – starts high, drops, and then rises

LML, HMH, MLM, MHM – (these are not very common.)

So that gives us a total of 15 different possible tones!  But as I said, languages will select from these.  As far as I can tell, the Mandarin tones are HH, LH, HL, HLH, and MM (the so-called neutral tone).

Fantasy Language Tone

Tone can be hard to do for a fantasy language, mostly because it is hard to put in a format that your publisher will accept!  If you decide to use a tone language, I would suggest using either two or four tones. 

The accents for tone go over the vowel.  If you have Microsoft Word, go to Insert -> Symbol and select “Combining Diacritical Marks.”  If you look at the selection of accents that comes up, you should be able to find the right accents for tone.

Represent low or falling tones with a grave accent: è à ì

Represent high or rising tones with an acute accent: é á í

Represent LHL tones with a circumflex: ê â î

Represent HLH tones with a breve accent: ĕ ă ĭ

For other tones—take a look at what your word processor has available, and pick from that.  Whatever you choose, be consistent, and keep track of what tones you have assigned to which words with which meanings!

Pitch-accent languages

Pitch-accent languages are complicated.  They share some of the characteristics of stress languages, and some of the characteristics of tone languages—they’re in between the two.

So how does that work?  Basically, a rule assigns stress to a particular syllable, based on number of syllables and syllable structure, as in a stress language.  However, the stress comes out in speech as a particular tone, usually a high tone of some kind.  AND—here’s the thing that makes pitch-accent so complicated—based on which syllable ended up getting stressed, tones will be assigned for all of the syllables in the word.

For example: (I’ll make up some rules…)

If the last syllable of a three syllable word is stressed (HH tone), both previous syllables receive LL tone.

If the second syllable of a three syllable word is stressed (HH tone), the first syllable receives LL tone, and the final syllable receives HL tone.

If the second syllable of a four syllable word is stressed (HH tone), the first syllable receives LH tone, the third syllable receives HLH tone, and the fourth syllable receives LL tone.

You can see that this gets complicated pretty quickly!

I don’t have much experience with pitch-accent languages, so I can’t give you any advice about them… except that, if you are serious about using a pitch-accent language, the best thing for you to do is probably to go and take a conversation class in Japanese.  This will help you to attune your ear to the pitch-accent system.