In addition to letters, the Shwa script includes punctuation. Like English punctuation, Shwa punctuation comes after the phrase it commands.


Instead of a space, Shwa uses a dot to separate words. This helps you read the separation between words as different from the separation between letters. In Abjad gait or topline fonts, a Dot is instead written as a simple interruption in the stroke linking the letters.

A Dot is one cell wide, like all the letters. Shwa fonts are fixed-width, and the glyphs are designed to be be of equal width (monospace). You don't need to break a line at the end of a word or to indicate a word that is split between lines ; just write letters all the way to the end of the line and then write the next letter on a new line. This means that lines of Shwa text are always both left- and right-justified, and "soft" newlines are not embedded in the text.

A blank line is used between paragraphs. The second consecutive Dot within Shwa text should be interpreted by the display (or replaced by the keyboard) as CR+LF(Carriage Return + Line Feed, U+000D + U+000A), in other words, a blank line. Chapters start on a new page: the third consecutive Dot indicates a new page (FF Form Feed, U+000C). So at the end of a chapter, the sequence will be "Dot CR LF FF".


In every language of the world, in addition to speaking with consonants and vowels, we also vary the loudness, rhythm and musical pitch of our speech to transmit information. Many languages use these suprasegmental mechanisms as stress, "accent" or tone, but those uses are lexical or morphological, part of the words themselves. In contrast, when they're applied to larger grammatical units, like entire phrases or clauses, we call them prosody. The most important element of prosody is pitch variation - melody - which we call intonation.

In written language, we try to indicate prosody with punctuation - the period, comma, question mark, exclamation point and so on - and also with typography : boldface, underlining or small caps. These mechanisms are all relatively new; even the space between words was only invented in the 7th century! And now we are inventing new punctuation at a rapid pace: not just the interrobang ‽, sarcmark , irony mark ⸮ and the like, but also emoticons like ☺ and abbreviations like lol.

This is the hardest part of Shwa, since we don't write intonation in our current scripts. We have to learn to listen to what we're actually saying, unconsciously, just as we did when first learning to write. But it's about time that we can finally write what we say - all of it!

Instead of these meaning-based symbols, Shwa offers a sound-based system of punctuation using the six accent marks without accompanying vowels. In other words, written Shwa communicates the suprasegmental features of language the same way that speech does: by indicating the intonation. For example, instead of using a ? question mark to indicate a question, Shwa writes the rising intonation of most questions. And in fact, that's more accurate: there are many exclamations that are phrased as questions ("What the heck are you doing?"), and many questions that are phrased as statements ("I bet you'd like some more tea."). In addition, intonation allows us to express much subtler shades of meaning: surprise, skepticism, displeasure.

Here's how it works in Shwa. The two Level accents, High and Low, are used to mark the end of intonation groups (also called tone groups, intonation units or intonation phrases), which correspond to information groups: each departs from information already known to the listener (the given) to add new information. An intonation group often, but not always, corresponds to a grammatical clause. The High Level accent is used for intonation groups that end on a rising tone, while Low Level is used when the group ends on a falling tone. They are written as suffixes attached to the final word in the group.

Within each intonation group is a tonic (or nucleus), which carries the most prominent pitch change. It's often, but not always, the last stressed syllable in the group, and it always represents the end of the new information - if anything follows, it's given. If it's not the last syllable, the pitch change is spread across the whole word from the nucleus to the end of the intonation group; it becomes a word melody - effectively, the whole word is the tonic.

In Shwa, tonics are marked with (one or more) slanted accents: High Rising, Low Rising, High Falling and/or Low Falling, always attached to the end of the word, even if the pitched syllable is earlier in the word. The High Rising accent is used for high rising tones, and so forth in a straightforward manner: the accent is interpreted as a small diagram of the pitch contour.

Did you see that? I saw it!

There are also tones with more complex patterns, and these are written with complex accents. A rise falling tone is an exaggerated rise followed by a lesser fall, as if to mollify the rise. It's written with a complex accent that has a High Rising Accent on top and a Low Falling accent on the bottom. Likewise, a fall rising tone is an exaggerated fall followed by a lesser rise, as if to mollify the fall. It's written with a complex accent that has a High Falling Accent on top and a Low Rising accent on the bottom. The first, dominant, accent is always on top.

You should do it. Why did they?

Sometimes, there is a second pitched element at the end of an intonation group, where there is a rise instead of the expected fall after a falling tonic. These compound tones are written in Shwa with only one slanted accent - the compound element is shown by the unexpected High Level accent at the end.

In other cases, there are other pitched elements before the tonic, and these pretonic heads can also be marked with slanted accents. Since the intonation group only ends with a Level accent, there's no ambiguity.

Much of the time, the tonic falls at the end of the intonation group, and Shwa makes this a little easier to write using final accents that combine the tonic accent with the intonation group accent, as follows :

Tone Tonic Accent As Final Sound
Low Falling
High Falling
High Rising
Low Rising
Fall Rising
Rise Falling

The final accents in the first and fourth lines are simply High and Low level accents, so that these two most common cases - low falling and low rising tones on the final word of the intonation group - are indicated by the absence of a separate accented tonic.

To recap, Shwa marks tonality (the division into intonation groups) using Level accents, tonicity (the position of the tonic) using slanted accents, and tone (the pitch contour) by choice of accent shape. To illustrate it, let's take a look at how this system can be used to punctuate English.

English Intonation

There are five nuclear tones in English, as described in Intonation in the Grammar of English [M.A.K. Halliday & William S. Greaves, 2008] :

  1. A Fall, usually written in Shwa with a Low Falling accent .
    However, there is a strong variant which is written with a High Falling accent .
  2. A Rise, written with a High Rising accent .
  3. A Level Rising, written with a Low Rising accent .
  4. A Fall Rising, written with a High Falling accent over a Low Rising accent .
  5. A Rise Falling, written with a High Rising accent over a Low Falling accent .

The second, third and fourth are followed by a High Level accent at the end of the intonation group. The first and fifth are normally followed by a Low Level accent at the end of the intonation group, but as mentioned above, there are also two compound tones :

  1. Fall + Level Rising, written in Shwa as a Low Falling accent on the tonic, with a High Level accent at the end of the intonation group: , then .
  2. Rise Falling + Level Rising, written in Shwa as a High Rising accent over a Low Falling accent on the tonic, with a High Level accent at the end of the intonation group: , then .

These tones amplify the meanings of English sentences, adding information about the flow of the discourse or the speaker's feelings about the content. For example, an unmarked declarative statement - one that carries no meaning beyond the meaning of its words - is spoken with tone 1. When spoken with tone 4, the tone adds a feature of reservation, as if there were a but involved somewhere. With tone 5, the additional meaning is one of surprise that the sentence is true. With tone 2, the sentence functions as a query, a challenge, or a response. With tone 3, the sentence is uncommitted: tentative, mildly agreeing, or simply incomplete, with the rest to follow. Tones 13 and 53 indicate a secondary focus.

Likewise, different tones add different shades of meaning to questions, queries (yes/no questions), responses, commands, offers, exclamations and vocatives. Intonation also plays an important role in amalgamating items into larger logical units, like conjunctions and lists. For instance, the sentence I thank my parents, the Pope and Sinéad O'Connor has two very different meanings depending on how it's pronounced. Halliday & Greaves go into great depth in describing how these tones are used to convey meaning, and I recommend you read the book to learn more. But to learn to punctuate Shwa, all you need is to be able to hear the different tones as you pronounce them, silently or out loud, and write what you hear.

Here's an example in the Latin alphabet, along with the audio, so you can focus on the punctuation:

(From Rudyard Kipling's Just So Stories)

In the beginning of years when the world was so new and all, and the Animals were just beginning to work for Man, there was a Camel, and he lived in the middle of a Howling Desert because he did not want to work; and besides, he was a Howler himself. So he ate sticks and thorns and tamarisks and milkweed and prickles, most 'scruciating idle; and when anybody spoke to him he said Humph!.

Lined Text

Normally, the final accent at the end of every intonation group is followed by a dot, and the text continues. That's what you type, and that's what's encoded. But display devices - screens and printers - often display them in a more sophisticated manner.

If vertical space is not an issue, for example a screen with scrolling, then the space after a final accent is usually replaced by a newline, so that each intonation group has its own line. Some of them are quite short, while others may wrap without a newline, but the result is to make it a little easier to parse each sentence.

Even in block text, the dot after a final accent is often followed with a wider space, so that intonation groups are still more separated than words within a group.

Defective Punctuation

In addition to the system just presented, Shwa offers a second, simpler system of punctuation for use when there isn't enough data for full punctuation. This might arise, for example, as a result of transcription of existing Roman text, as opposed to text written directly in Shwa.

In this defective system of punctuation, the period, question mark and exclamation point are simply transcribed using double accents :

Punctuation within a sentence uses single Level accents :

Defective punctuation is usually a "good first guess" for the correct full punctuation.


In the Roman alphabet, we use bold text, italics, underlining and different fonts to indicate when a passage of text should be read in a special way. Maybe it's a quote, or the words are being used as examples, not as part of the text.

In Shwa, we use letters with gaps in the middle, kind of like strikethrough without the line, and we call this style expanded. Expanded fonts usually omit the stems altogether, and modify the tops and bottoms. Here's an example:


Double accents are also used when entering outlines. Each item in the outline starts with a double accent whose shape indicates the relation of that item to the previous one :

For example:
  1. Mother
    1. Daughter
    2. Sister
      1. Child
      2. Sister
    3. Aunt
    4. Sister
      1. Daughter
  2. Great-Aunt

< Accents Gaits >

© 2002-2017 Shwa shwa@shwa.org 23may17