A summary of Sanskrit

If you know zero Sanskrit, you can still follow along with our series. But it certainly helps to have a basic sense of what Sanskrit is like and how it works. That's what we hope to give you here.

The alphabet

Sanskrit is written phonetically. Each sound has one symbol, and each symbol corresponds to one sound. The sounds below are provided in both the usual Devanagari script (संस्कृतम्) and in romanized Sanskrit (saṃskṛtam). All Devanagari in our lessons will be displayed next to its romanized version.

These sounds are colored according to where in the mouth they are pronounced. We will explain this system in the lessons to come.

  • a

  • ā

  • i

  • ī

  • u

  • ū

  • e

  • ai

  • o

  • au
  • अं
  • अः

  • ka

  • kha

  • ga

  • gha

  • ṅa

  • ca

  • cha

  • ja

  • jha

  • ña

  • ṭa

  • ṭha

  • ḍa

  • ḍha

  • ṇa

  • ta

  • tha

  • da

  • dha

  • na

  • pa

  • pha

  • ba

  • bha

  • ma

  • ya

  • ra

  • la

  • va

  • śa

  • ṣa

  • sa

  • ha

Given all of these sounds, we have the first question that the Aṣṭādhyāyī aims to answer: which phonetic distinctions are relevant to grammar?


In every spoken language, native speakers make subconscious changes to their speech so that they can speak more quickly and fluently. For example, some native English speakers might drop the final “g” of words like “running” or “drinking.” These kinds of changes are called sandhi.

Sanskrit sandhi changes are extensive, and they are almost always written down. These changes occur both within words and between words, and they depend both on specific sounds and on the semantics of different words and suffixes.

Given these sandhi changes, we have a second question: which sandhi changes apply in which contexts?

Basic words

Roughly, Sanskrit has three types of words. These are nominal words (nouns, adjectives, participles, and the like), verbs, and a broad third category we can call uninflected words. The example below uses each of these three word types:

  • रामो न जगाम।
    rāmo na jagāma.
    Rama didn't go.

Sanskrit also relies on something called inflection. Inflection is when we change part of a word to express a new meaning. English uses inflection in a limited way: we have one cat but two cats. Or perhaps you ate yesterday but will eat today. But Sanskrit nominals and verbs use inflection much more extensively:

  • नयसि
    You lead

  • नीयेरन्
    They might be led.

  • नेष्यताम्
    of those about to lead

  • निनीषन्तः
    those who want to lead

  • गजाय
    for the elephant

  • गजेषु
    among the (many) elephants

and in much more elaborate patterns, with multiple sandhi changes:

  • लभे
    I obtain.

  • रुणध्मि
    I obstruct.

This raises a third question: Which inflectional patterns apply in which contexts, and with what semantics?


Because Sanskrit words are highly inflected, Sanskrit does not usually depend on a specific word order. For example, the two sentences below have the same semantics:

  • रामो रावणं हन्ति
    rāmo rāvaṇaṃ hanti
    Rama kills Ravana.

  • रावणं रामो हन्ति
    rāvaṇaṃ rāmo hanti
    Rama kills Ravana.

Since word order is relatively unimportant in Sanskrit, the Aṣṭādhyāyī focuses instead on a fourth question: how do words with different semantics combine to express sentence-level semantics?

The human constraint

Finally, we should remember that the Aṣṭādhyāyī is part of a culture that values oral tradition and memorization. So a fifth question it tries to address is a pragmatic one: how can this system be compressed to the smallest possible form, so that it is easy to memorize and easy to recall?

With this basic framing, we are ready to begin.