In addition to the sounds described above, most languages use supersegmental features which vary the volume (loudness), musical pitch and timing of sounds. We'll describe them in that order.

Stress

English is a good example of a language with stress: some syllables are pronounced louder, at higher pitch and held longer than they otherwise would be. English words usually have one stressed syllable : the native stress falls on the first syllable of a word, but imported words often have stress on the next-to-last (penultimate) syllable. "Small" words like articles and common prepositions and conjunctions often have no stress, as if they were part of the surrounding words.

In Musa, stressed vowels are written high, hanging from the top of the line. To type them, you press the vowel key, followed by the Space key (while for low vowels you press Space before the vowel sign).

Here's how high vowels are used :

• In languages with no tone and fixed stress, like French, the vowel of the last full syllable in a stress group is written high, and other vowels are written low.
• In languages with a (free) stress accent, like English, Spanish or Russian, high vowels are used for stressed vowels, and unstressed vowels are written low.
• In languages with a pitch accent, like Japanese, high vowels are used for the last syllable before the fall in pitch, and other vowels are written low.
• In languages with tonal word accents, like Swedish and Norwegian, words with acute accent (accent 1 or single accent) use a high vowel on the first syllable, and words with grave accent (accent 2 or double accent) use a high vowel on both the first and second syllables.
• In languages with two register tones, like Navajo, high tones are written with high vowels, and low tones are written with low vowels.
• High pitch tones also cause vowels to be written high, as described below.

Secondary Stress and Metrical Notation

In many analyses, syllables are either stressed or not - there's no third possibility. But some scholars recognize secondary stress, a weaker form of stress.

To understand what's going on, we have to distinguish different sources of stress. The stress we discussed above - primary stress - is lexical: it's part of the pronunciation of the word. INsight and inCITE are both correct, but different; inSTANT is just wrong. Some other languages, like Chinese and French, don't have lexical stress, but they have syntactic stress. In French, the last word of each rhythmic group is stressed (a rhythmic group generally corresponds to a syntactic phrase).

On top of this primary stress is the tonic stress spoken on the nucleus of an intonation phrase. In the sentence My cat is crazy, the stress on the first syllable of crazy is stronger than on the word cat, since crazy is the new information. We mark that in Musa using Intonation marks, which you'll meet on the next page.

Below both of these is metrical stress, related to prosody and rhythmic feet. In English, we speak by alternating a stressed syllable with one or two unstressed syllables, as in The CAT in the HAT. If we have more than two unstressed syllables in a row, we tend to add a metrical stress to one of them, to "break up" the metrical foot. For example, in the word alphabetical, the primary stress falls on bet, but the first syllable is more stressed than the others. We normally show that stress by not reducing the al to ɐl, as we do in words like allow. And that's how we write it in Musa: as an unstressed (low) but unreduced vowel.

In other languages, different rules apply, but the idea is the same. For example, in many languages there's a distinction between heavy and light syllables (or long versus short): syllables with long vowels or final consonants are considered heavy, and attract secondary stress in the rhythmical system. A metrical foot contains one heavy syllable - one stressed syllable in English - and its associated light, short, or unstressed syllables.

However, Musa also offers a metrical notation for indicating prosodic meter when needed. With this notation, we indicate the boundaries of metrical feet with a lone Break  separated by spaces. The Break doesn't normally occur next to spaces, even in arithmetic notation (where it has many uses), so there's no ambiguity. When displaying meter like this, we also raise the vowels that are secondarily stressed - one per foot - so that every foot has a heavy syllable. Here's an example, the first couplet of Charivarius' The Chaos (1920), first in normal Musa and then in metrical notation:

Dearest creature in creation, Study English pronunciation.

  

  

You can see that, in addition to the Breaks indicating the boundaries of metrical feet (two of which, in this case, happen to replace "linking" offglides in the normal text), the poetic version raises the vowels of in and nun, which are not high in normal text, and writes them as unreduced.

Tones

Many languages also use tones to vary the musical pitch of vowels. Musa spells tones using small accent marks over or under the vowels. There are three of them :

Let me show you an example of how these accents are used to write tones. In Chinese, there are four contour tones, and a "fifth" tone consisting of no tone. Here are examples:

中文	Pinyin	Tone Number	Tone Description	Sound	Musa Tone Name	Musa
妈	mā	1st tone	Level tone		High Level tone	
麻	má	2nd tone	Rising tone		High Rising tone	
马	mǎ	3rd tone	Dipping tone		Low Level tone	
骂	mà	4th tone	Falling tone		High Falling tone	
吗	ma	"5th" tone	Neutral tone		No tone	

As you can see, the accents are written above low vowels to indicate a low tone, and below high vowels to indicate a high tone. It's the position of the vowel, not the accent, that makes a tone high or low. Also note that the level tone marks are used for peaking and dipping tones (which are usually written with a circumflex or caret ∧ and a caron or breve ∨).

Here's how this system extends to some other well-known tonal languages:

Tone	Musa	Language	Tone Name
Mid tone	 (high)	Cantonese	Upper Departing (3rd)
		Vietnamese	ngang (level)
		Thai	Mid
High Rising tone		Chinese	Rising (2nd)
		Cantonese	Upper Rising (2nd)
		Vietnamese	ngã (tumbling)
High Level tone		Chinese	High (1st)
		Hong Kong Cantonese	Upper Level (1st)
		Thai	High
High Falling tone		Chinese	Falling (4th)
		Guangzhou Cantonese	Upper Falling (1st)
		Vietnamese	nặng (heavy)
Low Falling tone		Chinese	3rd before a different tone
		Cantonese	Lower Level (4th)
		Vietnamese	hỏi (asking)
		Thai	Falling
Low Level tone		Chinese	Dipping (3rd)
		Cantonese	Lower Departing (6th)
		Vietnamese	huyền (hanging)
		Thai	Low
Low Rising tone		Chinese	3rd before another 3rd tone
		Cantonese	Lower Rising (5th)
		Thai	Rising
		Vietnamese	sắc (sharp)
No tone	 (low)	Chinese	Neutral ("5th")

Here's how the accents are used :

• In languages with contour tones, like Chinese, Cantonese, Thai and Vietnamese, the accents are used as shown in the chart above. Here's an example from Cantonese, first in fangzi gait, then Alphabet:

   

   

  我有一份工， 果份叫荫功。

  ngóh yáuh yāt fahn gūng, gwó fahn giu yām gūng.

• In languages with three register tones, like most Bantu languages, high tones are spelled with high vowels, low tones with low vowels, and mid tones with another accent, depending on the language. For example, the mid tone in Igbo is a downstep, written as a high falling tone, while the mid tone in Yoruba is written with a high level tone.
• In languages with four or more register tones, more accents are used, depending on the language. Musa has only 8 accents, while the world has many more tones, including those which include odd voicing. So Musa accents work well to distinguish tones within a language - no language I've seen has more than 8 tones - but not so well to spell the exact pitch and voicing consistenly between languages.
• In languages with tone sandhi - where tones change depending on the context - the tone may change based on the context. In such cases, the accent also changes. For example, when two Chinese syllables with Low Resting tone occur together, the first becomes Low Rising and the second one becomes Low Falling.
• In many tonal languages, the tones don't just involve pitch; they also involve contrasts in voice quality. The vowel might be voiceless, aspirated, slack, harsh, breathy, glottalized, stiff, hollow, creaky, or something else … or a combination of the above. In Vietnamese, for example, the huyền tone is breathy, the hỏi tone is harsh, the ngã tone is creaky, and the nặng tone ends sharply with a glottal stop. But Musa doesn't write any of this detail: the Musa tone accents indicate which (phonemic) tone applies, but not exactly how it's pronounced. In this respect, along with intonation, they depart from Musa's phonetic approach. But if you do need to write the acoustic details of tone contours, visit the discussion of orthophonic tones.

< Suffixes

Intonation >