This page brings together basic information about the Syloti Nagri script and its use for the Sylheti language. It aims to provide a brief, descriptive summary of the modern, printed orthography and typographic features, and to advise how to write Sylheti using Unicode.
Sylheti (ꠍꠤꠟꠐꠤ) is a language spoken by around 11 million people, mostly in eastern Bangladesh, but also in diaspora communities in the United Kingdom, the United States, Canada and the Middle East. The language is used as a first language by all in the ethnic community. It is not known to be taught in schools, but it is used and sustained by institutions beyond the home and community. The language has some spell checking or localized tools or machine translation as well.eth
Syloti Nagri (ꠍꠤꠟꠐꠤ ꠘꠣꠉꠞꠤ) is a South Asian abugida used in Bengal and Assam for the Sylheti language. It was primarily used in the eastern part of the Sylhet region, to document poetry known as puthis. In the course of the 20th century, it lost much ground to the standardised Eastern Nagari script, but since 2009 there has been an increased interest in the use of the script.
The Syloti Nagri script is an abugida, ie. each consonant contains an inherent vowel sound. See the table to the right for a brief overview of features for the Syloti Nagri script.
Text runs left to right in horizontal lines, but pages may run right to left. There is no case distinction. Words are separated by spaces.
Syloti Nagri represents native consonant sounds using 27 basic letters, which include 10 duplicates to indicate tone. Each consonant has a top bar that joins with those alongside. Syllable codas are written using ordinary letters or the anusvara combining mark. A special vowel killer is encoded to indicate word-final codas. There are no dedicated medial consonants.
Given that consonants normally include an inherent vowel, the orthography needs a way to indicate when a consonant is not followed by a vowel.
Consonant clusters may be unmarked (the Syloti Nagri Noto font only produces a few conjuncts) , or they may form conjuncts, or they may be indicated using a visible virama. The approach taken is very fluid, and may depend on the font or author preference. Conjuncts may be stacked or conjoined, but are the same height as normal letters. Inherent vowels are typically dropped at the end of a word, and the anusvara is used for nasal codas.
Plain post-consonant vowels are written using 5 combining vowel signs. A recent addition to these is a vowel that can indicate the inherent vowel sound when word-final. There are no composite vowel signs. Diphthongs are commonly written using a vowel sign followed by an independent vowel or, alternatively for those ending in -i, a dedicated combining mark can be used alone to represent ɔi or after a vowel sign for other diphthongs. There are no pre-base vowel signs or circumgraphs.
Standalone vowels are written using one of 5 independent vowel letters. In some texts, these vowel letters can carry a virama and combine with a following consonant.
Sylheti has a high tone, which is indicated by the consonant letter chosen.
Sylheti uses Bengali digits. Punctuation marks are mostly ASCII, but include the Devanagari danda and a small number of native symbols for poetry.
Notable features
high tone is indicated by a duplicate set of consonant letters
word-final vowel absence is not usually marked, but can be marked using a special, additional vowel killer
not all consonant clusters are marked
a vowel sign was recently added to indicate that a word-final consonant is followed by an inherent vowel sound
independent vowel letters may be combined with a following consonant letter using the virama
diphthongs can be written using a vowel sign followed by an independent vowel, or may involve a special diacritic for diphthongs ending in -i
page ordering may be right to left, although the text direction remains left to right.
some texts use 'false conjuncts' that require typed characters to not match pronunciation order.
The following represents the repertoire of the Sylheti language.
Click on the sounds to reveal locations in this document where they are mentioned.
Phones in a lighter colour are non-native, or allophones, or infrequent. Source Wikipedia.
Vowel sounds
Plain vowels
Complex vowels
io
Observation: Sylheti has several other instances where 2 (and sometimes 3) vowels occur together. Since those are written using a vowel sign followed by an independent vowel, it's unclear from the sources whether those constitute diphthongs or just sequences of vowel sounds.
Consonant sounds
labial
labio- dental
alveolar
post-
alveolar
retroflex
palatal
velar
glottal
stop
b
td
ʈɖ
kɡ
fricative
ɸ
f
sz
ʃ
x
ɦ
nasal
m
n
approximant
l
trill/flap
ɾ
ɽ
Tone
Sylheti distinguishes between normal and high/rising pronunciation of vowels, eg. compare:
cf.
ꠎꠣꠟ
ꠏꠣꠟ
Structure
tbd
Vowels
Vowel summary table
This table only summarises basic vowel to character assignments. Click on the phonetic transcriptions for more detail.
ⓘ represents the inherent vowel. Only the diphthong with a dedicated vowel sign is shown — other diphthongs can be written using a combination of vowel sign plus independent vowel (not shown here).
The inherent vowel for Syloti Nagri is pronounced ɔ. So bɔ is written by simply using the consonant letter.
eg.
ꠛꠞꠔꠘ
ꠛ,ꠞ,ꠔ,ꠘ
Since Sylheti consonants normally include an inherent vowel, the orthography has ways to indicate a consonant that is not followed by a vowel sound. See novowel.
Post-consonant vowels
Plain post-consonant vowels are written using 5 combining vowel signs. There are no composite vowel signs.
Diphthongs are commonly written using a vowel sign followed by an independent vowel or, alternatively for those ending in -i, a dedicated combining mark can be used alone to represent ɔi or after a vowel sign for other diphthongs.
Three vowel signs are spacing marks, meaning that they consume horizontal space when added to a base consonant.
Sylheti uses the following dedicated combining marks for plain vowels.
ꠤ,ꠥ,ꠦ,ꠧ,ꠣ
ꠧ is a recent addition to the orthography, driven, in particular, by a desire to differentiate a word-final o vowel in imperative forms of certain verbs, and designed to be similar to signs used in other scripts. It is not attested in historical texts.d,12 It is also used to write ɔ at the end of a word, since a word-final inherent vowel is typically not pronounced. For example, compare:
cf.
ꠝꠣꠗꠣꠘ
ꠇꠥꠘꠣꠘꠧ
Diphthongs
ꠇꠂꠣ
xai
ꠇꠣꠁ
xai
The sources consulted blur the lines between representation of diphthongs versus post-consonant vowels followed by standalone vowels. (In Hindi written with Devanagari these represent different sounds.) But it seems that this blurring is sustained by the orthography, since words with the same underlying pronunciation can be spelled in both ways.
Sylheti has one dedicated vowel sign used to represent diphthongs, called dvisvara.
ꠂ
Used alone, it creates the diphthong ɔi~oi, which is relatively common.
eg.
ꠞꠂꠖ
ꠄꠇꠂꠡ
However, less commonly, it can also be used to create other diphthongs ending in -i by adding it after another vowel sign.d,11-12
eg.
ꠀꠂꠎꠇꠂꠣꠟ
ꠖꠂꠥ
Lloyd-Williams et al indicate that when combined with a spacing vowel sign, such as ꠣ, ꠂ may be rendered over the consonant or over the vowel-sign.d,12 The Noto Sans Syloti Nagri and Surma fonts only support the order of code points { Consonant + Dvisvara + Vowel_sign }, and put the dvisvara above the consonant, ie.
ꠇ,ꠂ,ꠣ,ꠇꠂꠣ
However, such diphthongs are also commonly written using independent vowels, rather than with dvisvara.
ie.
ꠖꠥꠁ
ꠖꠥꠁ,ꠖ,ꠥ,ꠁ
ꠀꠁꠎꠇꠣꠁꠟ
ꠀꠁ,ꠎ,ꠇꠣꠁ,ꠟ
Other diphthongs (ie. not ending in -i) are constructed in the same way.
eg.
ꠇꠌꠥꠀ
ꠇ,ꠌꠥ,ꠀꠤ
ꠇꠤꠀꠘꠧ
ꠇꠤꠀ,ꠘꠧ
Standalone vowels
ꠅ
ɔ
Syloti Nagri uses one of the following independent vowel letters to write standalone vowels.
ꠁ,ꠃ,ꠄ,ꠅ,ꠀ
eg.
ꠀꠋꠉꠥꠞ
ꠃꠜꠣꠘꠤ
ꠁꠟꠣ
Word-medial independent vowels are common, and may create a sequence of vowel sounds.
eg.
ꠇꠣꠃꠀꠟꠥꠟꠤ
ꠇꠣ,ꠃ,ꠀ,ꠟꠥ,ꠟꠤ
ꠇꠦꠅꠀꠟꠣ
ꠇꠦ,ꠅ,ꠀ,ꠟꠣ
Vowel-consonant ligatures
Lloyd-Williams et al describe independent vowels ligating with following consonants, although it is not very commond,21. The ligatures can be activated in the Surma font using the hasant.
eg.
ꠀ,꠆,ꠔ,ꠀ꠆ꠔ
Observation: This behaviour cannot be reproduced using the Noto Sans Syloti Nagri font.
The following are a few examples of ꠀ ligating with a following consonant.
ꠀ꠆ꠔ,ꠀ꠆ꠡ,ꠀ꠆ꠘ,ꠀ꠆ꠞ,ꠀ꠆ꠟ
Tones
Syloti Nagri indicates whether a vowel should be pronounced as normal or with a high/rising tone by preceding the vowel with the appropriate symbol for the consonant sound.
eg.
ꠎꠣꠟ
ꠎ,ꠣ,ꠟ
ꠏꠣꠟ
ꠏ,ꠣ,ꠟ
Not all consonant letters have these alternate forms. Those that are are letters that were historically associated with aspirated stops (it is thought that the tone is a result of aspiration becoming lost through historical sound changes). The following letters have no high tone equivalent.
ꠡ,ꠢ,ꠝ,ꠘ,ꠞ,ꠠ,ꠟ
Syloti Nagri also appears to use an apostrophe to indicate high tone sounds when no consonant is available.
eg.
ʼꠅꠄ
ꠙʼꠅꠞ
Observation: It's not clear which code point should be used for this. In the example here we use ʼ. It's also not clear whether the apostrophe should appear to the left or right of the letter with the high tone.
Vowel sounds to characters
This section maps Sylheti vowel sounds to common graphemes in the Syloti Nagri orthography.
Sounds listed as 'infrequent' are allophones, or sounds used for foreign words, etc. Light coloured characters occur infrequently.
Plain vowels
i
vowel signꠤ
independentꠁ
u~ʊ
vowel signꠥ
independentꠃ
e
vowel signꠦ
independentꠄ
o
vowel signꠧ
independentꠅ
ɛ
vowel signꠦ
independentꠄ
ɔ
inherent vowelⓘeg. ꠉꠞꠝ
vowel signꠧUsed word-finally to indicate the presence of a final ɔ sound.
independentꠅ
æ
vowel signꠦ
independentꠄ
a
vowel signꠣ
independentꠀ
Diphthongs & rhymes
ɔi~oi
ꠂ
Vowel absence
Vowel absence principally occurs either when a consonant is a syllable coda, or when a consonant is part of a consonant cluster.
Inherent vowels are typically dropped at the end of a word.
eg.
ꠀꠁꠎ
ꠉꠎꠟ
Consonant clusters are also commonly unmarked.
eg.
ꠛꠞꠔꠘ
ꠇꠤꠍꠝꠤꠍ
ꠔꠣꠘꠔꠣꠘ
Note that the last example above could have been written using a conjunct (ie. ꠔꠣꠘ꠆ꠔꠣꠘ), but wasn't.
Conjuncts
Conjunct formation
To produce a conjunct, ꠆ is added between the consonants in the cluster. When a conjunct is formed, the virama itself is not displayed.
eg.
ꠞ,꠆,ꠟ,ꠞ꠆ꠟ
Conjuncts can, in principle, involve any combination of characters in Syloti Nagri, but in practice their frequency is fairly limited. The Noto Sans Syloti Nagri only produces 17 conjuncts for 2-consonant combinations. On the other hand, the Surma font produces 155 conjuncts for 2-consonant clusters.
Conjuncts are not used across morphological boundaries, and, although there are ready exceptions, they are generally avoided for Arabic and Persian loan words, however they are commonly found in English loan words.d,20
Conjuncts are typically formed by either stacking or conjoining glyphs.
Stacking
The initial and second consonants are combined vertically, but the overall height of the conjunct is the same as a normal letter. For example:
ꠇ,꠆,ꠔ,ꠇ꠆ꠔ
ꠌ,꠆,ꠔ,ꠌ꠆ꠔ
ꠛ,꠆,ꠛ,ꠛ꠆ꠛ
Examples of conjuncts formed by subjoining non-initial consonants.
Conjoined consonants
Consonants are merged, side by side. Where the initial consonant has a vertical line to the right, this is typically removed to create a half-form.
ꠘ,꠆,ꠖ,ꠘ꠆ꠖ
ꠡ,꠆,ꠇ,ꠡ꠆ꠇ
ꠟ,꠆,ꠙ,ꠟ꠆ꠙ
Examples of conjuncts formed by subjoining non-initial consonants.
'False' conjuncts
Lloyd Williams et al describe certain words where a conjunct is formed even though a vowel is pronounced between the two consonants. This sometimes appears to be related to sound changes that are not reflected in a spelling based on Bengali orthographies, but such conjuncts were apparently also used to manage line length in poetry.d,23-26 This is not common in modern printed text.
For example, the word:
ꠁꠝꠣꠘ
ꠁ,ꠝ,ꠣ,ꠘ
may be written:
ꠁꠝ꠆ꠘꠣ
ꠁ,ꠝ,꠆,ꠘ,ꠣ
The order of characters for these 'false' conjuncts is expected to follow the spelling, rather than the pronunciation.
Visible vowel killer
The Syloti Nagri block has two code points that can be used to explicitly indicate inherent vowel suppression. In some consonant clusters the first converts the cluster to a conjunct (see clusters), but it is often seen as a visible marker.
Hasant
The hasant, ꠆, appears as a fallback when a font doesn't contain conjunct glyphs for a consonant cluster. However, because fonts may only contain a small number of conjuncts for Syloti Nagri, it may also appear in certain words as a visible mark (such as in the examples below). It wasn't widely introduced to the orthography until the mid 20th century.
eg.
ꠚꠣꠟ꠆ꠇꠥꠘ
ꠚꠣ,ꠟ꠆,ꠇꠥ,ꠘ
ꠇꠝ꠆ꠟꠣ xɔmlaorange
ꠇ,ꠝ꠆,ꠟꠣ
The first example above uses the Surma font. The second uses the Noto Sans Syloti Nagri font, which contains far fewer conjunct glyphs — in the Surma font the hasant would not be visible for that word.
Word-final hasant
Word-final consonants are normally not followed by an inherent vowel sound. However, occasionally an author wants to make it explicit that the inherent vowel is dropped. Because, in some fonts, ꠆ when rendered would project too far beyond the final coda, authors may instead use a mark below the coda, shaped like the Bengali hasant. ꠬ is designed for this purpose. Its use is rare.lpe
This hasant is never used for consonant clusters, and therefore never creates conjuncts. It is always visible.
eg.
ꠇꠞ꠬ !
Consonants
Consonant summary table
This table only summarises basic consonant to character assignments. Click on the phonetic transcriptions for more detail.
The left-hand column shows ordinary letters for stops and fricatives. The right-hand column shows those letters that produce a following, high tone vowel sound.
The following set of consonant letters cover the basic consonant onset sounds.
Click on each letter for usage notes, alternative pronunciations, and for examples of usage.
ꠙ,ꠛ,ꠔ,ꠌ,ꠖ,ꠎ,ꠐ,ꠒ,ꠇ,ꠉ,ꠡ,ꠢ,ꠝ,ꠘ,ꠞ,ꠠ,ꠟ
A further set of consonant letters change the tone of the following vowel.
ꠚ,ꠜ,ꠕ,ꠍ,ꠗ,ꠏ,ꠑ,ꠓ,ꠈ,ꠊ
For example, compare the two following words:
cf.
ꠔꠣꠟ
ꠕꠣꠟ
Onsets
Consonant clusters don't normally occur in Sylheti onsets.d
Codas
Most codas are written using normal consonant letters, but Syloti Nagri uses the anusvara combining mark for ŋ codas.
Examples of the use of ordinary consonants that don't mark the consonant clusters within a word.
eg.
ꠀꠇꠟꠖꠣꠔ
ꠀ,ꠇ,ꠟ,ꠖꠣ,ꠔ
ꠀꠙꠘꠦ
ꠀ,ꠙ,ꠘꠦ
Some words may use conjuncts (see clusters), or visible vowel killers (see visible_virama).
eg.
ꠃꠘ꠆ꠗꠣꠟ
ꠌꠤꠘ꠆ꠔꠣ
ꠚꠣꠟ꠆ꠇꠥꠘ
Anusvara
The sound ŋ occurs only at the end of a syllable, and is written in Syloti Nagari using the combining mark ꠋ.
When a consonant is followed by a spacing vowel sign, the anusvara most frequently appears above the consonant in manuscripts, but in digital texts it most frequently appears above the vowel sign, and it is expected that this will the norm going forward.d,13 The expected code point order is { Consonant + Vowel_sign + Anusvara }.
This mark can appear over consonants, over vowel signs, and over independent vowels. The following examples show it respectively above an independent vowel, a spacing dependent vowel, a non-spacing dependent vowel that occupies the same space, and a consonant (with inherent vowel):
eg.
ꠀꠋꠑꠤ
ꠉꠣꠋ
ꠛꠦꠋꠉꠟꠤ
ꠞꠋ
Consonant length
Geminated consonants are common in Sylheti. They are written using conjuncts or with a visible hasant, depending on the font capablities.
eg.
ꠃꠘ꠆ꠘꠤꠡ
ꠇꠐ꠆ꠐꠣ
ꠉꠥꠟ꠆ꠟꠤ
Observation: Of the 17 conjuncts supported by the Noto Sans Syloti Nagri font, 7 are for geminated consonants.
Consonant sounds to characters
This section maps Sylheti consonant sounds to common graphemes in the Syloti Nagri orthography.
Sounds listed as 'infrequent' are allophones, or sounds used for foreign words, etc. Light coloured characters occur infrequently.
b
ꠛ
ꠜProduces a high tone vowel after.
t̪
ꠔ
ꠕProduces a high tone vowel after.
d̪
ꠖ
ꠗProduces a high tone vowel after.
ʈ
ꠐ
ꠑProduces a high tone vowel after.
ɖ
ꠒ
ꠓProduces a high tone vowel after.
k
ꠇUsually pronounced x, but pronounced as k before and after i or u, or when geminate.
ꠈProduces a high tone vowel after. Also usually pronounced x, but pronounced as k before and after i, or u or when geminate.
ɡ
ꠉ
ꠊProduces a high tone vowel after.
ɸ~f
ꠙ
ꠚProduces a high tone vowel after.
s
ꠌ
ꠍProduces a high tone vowel after.
z
ꠎ
ꠏProduces a high tone vowel after.
ʃ
ꠡ
x
ꠇ
ꠈProduces a high tone vowel after.
ɦ
ꠢ
m
ꠝ
n
ꠘ
ŋ
ꠋCoda.
ʊ~w
ꠅWhen at the end of a syllable.
ɾ
ꠞ
ɽ
ꠠIn ordinary speech this is often pronounced ɾ.
l
ꠟ
Encoding choices
This section offers advice about characters or character sequences to avoid, and what to use instead. It takes into account the relevance of Unicode Normalisation Form D (NFD) and Unicode Normalisation Form C (NFC).
Although usage is recommended here, content authors may well be unaware of such recommendations. Therefore, applications should look out for the non-recommended approach and treat it the same as the recommended approach wherever possible.
Codepoint sequences
Combining marks always follow the base character.
ꠋꠋ is typed and stored after any dependent vowel sign.
When ꠂ is used in conjunction with a dependent vowel sign, the Noto and Surma fonts require that it precede the vowel sign. It's not clear whether this is the preferred order or not.
Numbers
Digits
Despite some experiments, contemporary Sylheti doesn't have a set of native, decimal digits. It uses Bengali digits, instead.d,19
১,২,৩,৪,৫,৬,৭,৮,৯,০
Currency
The currency symbol is ৳d,18.
Text direction
Syloti Nagri text runs left to right in horizontal lines, but pages sometimes run right to left.
Lloyd-Williams et al suggest that left-to-right page sequencing is an appropriate default for modern printed text, but that right-to-left sequencing be available for certain kinds of book, such as historic religious poetry.d,41
Syloti Nagri requires both context-sensitive shaping and positioning. Multiple combining marks can attach to a single base letter.
Cursive or joining letters
The script is not cursive in the sense that Arabic is, but the top bar above each consonant letter joins with those on either side. This can have an impact for highlighting strokes, or for letter-spacing.
Typographic units
Word boundaries
Words are separated by spaces in modern, printed Syloti Nagri text, though not in historical texts.d,36
Graphemes
tbd
Punctuation & inline features
Phrase & section boundaries
For normal text, Syloti Nagri uses mostly ASCII punctuation marks, but includes dandas from the Devanagari block. Poetry uses some additional, Syloti Nagri specific, punctuation.
Syloti Nagri commonly uses ASCII parentheses to insert parenthetical information into text.
start
end
standard
(
)
Quotations & citations
Lloyd-Williams et al indicates that Syloti Nagri uses quotation marks like the following. Of course, due to keyboard design, quotations may also be surrounded by ASCII double and single quote marks.
start
end
initial
“
”
Line & paragraph layout
Line breaking & hyphenation
Lines are generally broken between words.
Hyphenation is not normally seen in modern texts, but Lloyd-Williams et al argue that it would be useful in some circumstances, in which case hyphenation opportunties would occur before any consonant or independent vowel (presumably, except inside conjuncts).d,36
Text alignment & justification
It is traditional for Syloti Nagri texts to be fully justified.d
Page & book layout
General page layout & progression
Syloti Nagri books may order pages right to left, even though the text direction remains left to right. This is most common for books on poetry.
