Brahmi Inscriptions

The Kushan Brahmi Alphabet

It is the considered opinion of specialists in paleography, such as Salamon & Dani, that there is no such thing as 'Kushan' Brahmi. They argue that letter forms vary from site to site and over time, but that this does not correspond with the rule of particular dynasties. So that terms such as 'Ksatrap', 'Gupta', or 'Kushan' Brahmi do not make sense. This is a reasonable point, and it would be more accurate to describe the Brahmi presented here as '2nd & 3rd Century Mathuran' Brahmi since it is the form that tended to be found in Mathura in the second and third century AD. However, this is a cumbersome way to work, so the term Kushan Brahmi is used here, to refer to the sequence of inscriptions from Mathura belonging (mostly) to the period between Kanishka I and Vasishka.

Basic Principles

The alphabet consists of basic symbols (aksaras) which represent open syllables (such as 'ka', 'ta', or 'bha'). See the first table below, which shows all the symbols for all the syllables that end in 'a'. When writing a syllable which ends with a different vowel the aksara is modified. The second table shows how each vowel modifies a letter with a series of examples. It also shows the symbol which is used when a vowel begins a word, which are the only occasion on which a vowel stands alone as a separate symbol. When Brahmi is used to indicate a cluster of consonants, the symbols for both consonants (and in some cases three consonants) are combined together to form a ligature. Some of the more common ligatures are shown in the third table. Lastly, the fourth table shows the Brahmi numerals, though it should be noted numerals are often very variable in their form.

All of the symbols here are based directly on readings of inscriptions, from the region of Mathura, and from the first to the thrid century AD. For the simple reason that not every character occurs in the Kushan inscriptions, not every character is shown in these tables. They will therefore differ tables presented by other authors, most of which tend to focus on the Asokan period. For a series of tables indicating letter forms from a wide variety of regions and periods readers should consult Dani (1986).


Latin characters are displayed in these sections using the Lucida Sans Unicode font. If the list of characters below are not displaying correctly (usually boxes appear) you will need to download and install this font by right clicking on this link and choosing save as, then saving the file to the fonts directory in windows.

a ā i ī u ū ṛ ṝ ḷ e ai o au k kh g gh ṅ c ch j jh ñ ṭ ṭh ḍ ḍh ṇ t th d dh n p ph b bh m y r l v ś ṣ s h ṁ

1. Basic Characters (aksara)

ka kha ga gha ṅa n/a
ca cha ja jha ña  n/a ya śa 
ṭa ṭha ḍa ḍha n/a ṇa  ra ṣa
ta tha da dha na la sa
pa pha n/a ba bha ma va ha

2. Vowel Modification

Vowel a i ī  u e o
Initial Form  
Modification n/a or or oror
Example (in k)    
Example (in y)        
Example (in r)    

3 Ligatures

tra tsa rva śya
ṣka sya ryya tta
pra ṣṭhā  kṣu    

4 Numerals

1 2 3 4 5
6 7 8 9    
10 20 30 40 50
60 70 80 90 100 n/a


ṣṭhā A.B. A shows how much the letter form can vary. B stands in the same place as this ligature normally occurs and is normally transliterated in the same way but it may be another letter 'stha'.
  Dani (1986:89) notes that at Mathura it is common for the anusvara to become a short line above the character instead of a dot.
ṇa & na   The forms of these two letters can become confused in inscriptions to the degree that they become almost indistinguishable. 
ṣa A. B. Though the aksara is uncommon in its own right it is commonly found as a ligature ṣṭhā or kṣu. There are two forms, the open (A) and closed (B). They are interchanged not only at the same period but even in the same inscriptions.
kṣu This is the open form.
ña & ṅa   The character is not shown in the character table for the simple reason it does occur in Kushan inscriptions. It is in fact very rare in any inscriptions (32, 69, 89, 165) and is part of a ligature
jha The character is infrequent and therefore readings of it are open to some doubt.
pa, la, ha These three characters are similar and easily confused. La sometimes tends for the right vertical to be higher than the left. Ha tends for the right vertical to curl over.
da This character is subject to considerable variation, probably the results of attempts to write it in a more cursive manner.
40 & 70 & Though seemingly distinct in the tables above numerals are subject to such wide variation that it is a very simple matter to confuse these two. Any date involving 70 or 40 should be subject to serious doubts.
100   There is no character for 100 shown. Dates in the Kushan period do not include it, and any character read as 100 should be subject to serious doubts.

Contents Page Brahmi Inscriptions



Inscription List


Paleographic Dating




Robert Bracey