I recently got this question from an unknown source (it came to me through a friend of a friend of a friend route)—

In Hindi, the verb is inflected for gender, e.g. “वह खाता है” (He eats) vs. “वह खाती है” (She eats). This seems to also be the case for Marathi, “तो खातो” (He eats) vs. “ती खाते” (She eats). As a native speaker I can attest that to also be the case in Punjabi. However, in their root language, Sanskrit verb isn’t inflected for gender. Interestingly, as we see from above examples, in English, too, the verb doesn’t change based on the gender - that ostensibly isn’t rooted in Sanskrit but, say, Latin. I don’t know other European languages to offer any other data point. Am I right in these observations? If so, then there doesn’t seem to be a pattern as to which languages would inflect the verb for gender and which won’t? Looks like this trait evolved independently of a language’s roots. Thoughts?

The observation is correct. Gendered verb forms are found in most Indian languages, yet it’s true that gender is not marked on Sanskrit verb forms.

It’s true that finite verb forms in Sanskrit are not marked for gender. By that I mean words like पठति (he / she reads), गच्छामि (I go) etc., which are referred to in Paninian grammar as तिङन्त forms, as they end in the finite-verbal तिङ् suffix (प्रत्यय). If you have studied Sanskrit, this is the 3x3 matrix of verb forms (धातु रूप) you memorized (पठति पठतः पठन्ति etc.).

However, Sanskrit also uses non-verbal forms for conveying similar or related semantics. By non-verbal forms, I am referring to words that are analyzed as noun-like entities in the Paninian framework and referred to as सुबंत forms, as they end in the nominal सुप् suffix (प्रत्यय). This is the 8x3 matrix of noun forms you memorized (रामः रामौ रामाः etc.).

Examples of these nominal forms—

  • Present participle forms generated by adding the शतृ / शानच् प्रत्यय, such as गच्छन् / वर्धमानः, meaning going and growing respectively.
  • Past participle forms generated by adding the क्त / क्तवतु प्रत्यय, such as गतः / गतवान्, both meaning gone.

These are often used as adjectives in Sanskrit. In Western linguistic terminology, these are usually referred to as participles, and analyzed as verbal-adjectives. Like other noun-forms in Sanskrit these are also gendered.

Compare—

1a. सः गतवान्। (he went)

2b. सा गतवती। (she went)

The most likely hypothesis for the route through which modern Indian languages ended with gendered verbs is via Prakrit, which is classified as a Middle Indo-Aryan Language. Prakrit had a significantly simplified tense system compared to the 10 Sanskrit लकार, typically retaining only 4 (in modified form), and none of the past tense forms were retained. Instead participle forms such as the ones given above became much more common.

However, Indian languages aren’t the only ones that show gender marking on verbs. Semitic languages like Arabic or Hebrew have a similar system with verbs marked for gender. Russian has its past tense forms marked for gender, interestingly for the same reason as Prakrit—they originated from participle forms.

While we are on the topic, another interesting feature related to gendered verbs in Indian languages is the shift in the verb’s gender agreement depending on the construction. If I ask you to guess the entity whose gender and number is marked on the verb, I suspect you might guess ‘the subject’, but consider two nouns आम (mango, masculine), and मिर्ची (chilli, feminine), and the following sentences—

2a. राम आम खाता है। (Ram eats a mango)

2b. राम मिर्ची खाता है। (Ram eats a chilli)

3a. सीता आम खाती है। (Sita eats a mango)

3b. सीता मिर्ची खाती है। (Sita eats a chilli)

Here, the verb matches the gender and number of the subject (Ram or Sita) and is agnostic of the object.

Now consider instead—

4a. राम ने आम खाया। (Ram ate a mango)

4b. राम ने मिर्ची खाई। (Sita ate a mango)

5a. सीता ने आम खाया। (Sita ate a mango)

5b. सीता ने मिर्ची खाई। (Sita ate a chilli)

The verb is still marking the gender, but now it’s matching the gender of the object and is agnostic of the subject!

This pattern called split-ergativity, is found in multiple Indian languages. The origin of this is also postulated to be the slow transformation of the Sanskrit passive construction using क्त प्रत्यय into the modern form.

Consider the strong similarities between 4a and 4b with 6a and 6b—

6a. रामेण पत्रम् लिखितम्। (A letter was written by Ram)

6b. रामेण कविता लिखिता। (A poem was written by Ram)

References:

  1. The Indo-Aryan Languages, Edited By George Cardona, Danesh Jain, link.
  2. Grammaticalization of participles and gerunds in Indo-Aryan: Preterite, future, infinitive by Annie Montaut, link.
  3. Ergativity in Indo-Aryan, Pritty Patel-Grosz, link.
  4. Typological Variation in the Ergative Morphology of Indo-Aryan Languages, Ashwini Deo and Devyani Sharma, link.