A gerund is basically a term given to the process of generating a meaning of “something having done”. A gerund is an indeclinable participle (i.e. it has no person or number) as in it acts as an adverb that modifies the main verb of the sentence. A gerund is formed by adding the suffix त्वा/इत्वा or य directly onto the verbal root.

For example, if we have to say that Arjuna sees an elephant after walking into the jungle. The basic way of saying it will be by using ca.

अर्जुनो वनं चरति स च गजं पश्यति
arjuno vanaṃ carati sa ca gajaṃ paśyati
Arjuna walks to the forest, and he sees the elephant.

However, the sentence is not sounding proper as it is not providing any concrete connection between Arjuna walking in the forest and seeing the elephant. So something needs to be changed. Let us try with some modifications.

अर्जुनो वनं चरित्वा गजं पश्यति
arjuno vanaṃ caritvā gajaṃ paśyati
After walking to the forest, Arjuna sees an elephant

Now the sentence seems to be complete and of logical sense. The change here is called gerund. By adding त्वा as a suffix we have managed to alter the sense and make it more meaningful. This is what in short gerund does.

A gerund is like an uninflected verb, which has action and can also interact with the nouns. It has almost all the qualities of a regular verb.

Again one thing to note is that only Sanskrit, of all the Indo-European language, has the concept of something called gerund but just like iti and tva, gerund seems to be borrowed from the Dravidian language.

What makes gerund tricky is that it is easy to understand but difficult to produce. However, if one wants to produce gerund, they can do so by following two simple rules i.e.

1. by weakening the roots and
2. By adding a suffix. Gerund uses two sandhis; ya and tva. In order to add a suffix, one needs to be thorough with the rules of sandhi.

Let us take a look at gerund with each suffix

Gerund with ya suffix:

उपसं-गम् → उपसं-गम्य
upasaṃ-gam → upasaṃ-gamya
come → having come

सम्-इ → सम्-इत्य
sam-i → sam-itya
come together → having come together (we haven’t studied this verb yet)

We have seen in earlier part that we attach ya to the roots, but when it is a short vowel we attach tya instead of ya. We can, however, weaken the root to the short vowel, long vowel or else we can keep it same.
Gerund with tva suffix:

तॄ → तीर्त्वा
tṝ → tīrtvā
cross → having crossed

पॄ → पूर्त्वा
pṝ → pūrtvā
fill → having filled

The tva rule is little difficult to understand as the suffix changes from root to root. Some grammarians use itva instead to make gerund sound easier, however, it is not a strict rule to do so, but it becomes easier to use once we are thorough with the sandhi rule. The rule that we have used in above example is the r rule. The r sound changes to become ir in almost every case. It is a rare vowel, hence in the course of time, the vowel has become to *tṝtvā and *pṝtvā.

Another rule in Sanskrit is about internal sandhi rule.

If the root ends in a voiced aspirated letter (gh, jh, ḍh, dh, bh), then the t- of the suffix becomes dh and the last letter of the root loses its aspiration. In other words, a voiced and aspirated consonant shifts its voice and aspiration forward.

बुध् + त्वा → बुद्ध्वा
budh + tvā → buddhvā
having awakened

लभ् + त्वा → लब्ध्वा
labh + tvā → labdhvā
having obtained