As we have learned, Sanskrit text—whether in metrical verse, open verse, or general prose—is composed of words, a significant portion of which are generated using stems and roots, following systematic rules of word construction. All words in Sanskrit can broadly be classified into two categories: those that remain unchanged regardless of grammatical factors, known as avyaya (indeclinable words), and those that inflect based on person (first, second, or third), case (Nominative, Accusative, Instrumental, Dative, Ablative, Genitive, and Locative), number (singular, dual, or plural), tense or mood, and gender.
Starting with inflected words, we have nouns (naama-pada), which, like in any other language, refer to living beings, objects, and places. Every noun in Sanskrit has an uninflected noun-stem (praatipadika), but unlike in many Indian languages, this base form is never used in speech or writing. For example, the name Ravi appears in its base form only in the vocative case when directly addressing someone. In all other cases, a suffix is added, which changes based on the noun’s ending vowel, gender, and grammatical case (vibhakti).
Sanskrit nouns always belong to one of three grammatical genders—masculine, feminine, or neuter. While many nouns have an obvious gender, others require suffix modifications to derive their masculine, feminine, or neuter forms. For instance, Krishna (m.) and Krishnaa (f.), Guru (m.) and GurvaaNi (f.), Yogi (m.) and Yogini (f.), Pitaamaha (m.) and Pitaamahi (f.), Saadhu (m.) and Saadhvee (f.).
Sanskrit employs a system of seven grammatical cases (vibhakti), and because nouns also distinguish singular, dual, and plural forms, this results in a 3 × 7 = 21 case-ending structure. While Sanskrit follows a general template for these 21 case endings, pronunciation constraints led to further variations depending on the ending vowel in the noun-stem, its long and short forms, and its gender. Since noun-stems with the same vowel endings can belong to different genders, additional tables are required to account for these distinctions. Although students eventually recognize patterns within vibhakti tables, mastering noun inflections still requires memorizing nearly twenty different sets of vibhakti tables.
Sanskrit also has a more extensive set of pronouns than many other languages. Personal pronouns (“I,” “You,” “He/She/It”) are similar to those found in most languages, but demonstrative pronouns (“this” and “that”) also have distinct masculine, feminine, and neuter forms in Sanskrit. Notably, the pronoun “this” has two variations: etat and idam. In addition to personal and demonstrative pronouns, Sanskrit includes interrogative, relative, and indefinite pronouns, all of which take gendered forms. Since pronouns function as substitutes for nouns, they follow vibhakti inflection rules as well. However, once a pronoun is known, deriving its inflected forms is relatively straightforward, as they follow the same noun vibhakti patterns based on their stem ending and gender.
Adjectives in Sanskrit, whether descriptive, locational, or numerical, also get modified and inherit the linga (gender), vachana (number), and vibhakti form of the naama-pada (noun) they are associated with. Thus, in Sanskrit adjectives, we see modifications like these: dvau dushṭau baalaau bheetaaṁ gaaṁ anudrutaḥ (“The two naughty boys chase the scared cow”). Pronoun adjectives like sarva (all), kiyat (few), katara (which), etc., do not follow these rules and have their own vibhakti tables referred to in the previous paragraph. Given the fact that many naama-pada in Sanskrit also function as adjectives, it can give rise to ambiguities—e.g., a stand-alone sentence like guruḥ shishhyaḥ, because guru is both a naama-pada (the teacher) and an adjective (heavy), can mean both “the heavy student” or “the teacher is the student,” which brings out the context-sensitivity of the Sanskrit language.
The next most important part of speech in Sanskrit is the kriyaa (verb). As previously mentioned, Sanskrit verbs originate from dhaatu (verb roots), which serve as the foundation not only for kriyaa but also for derived naama-pada, kriyaavisheshaNa (adverbs), and adjectives. To form a kriyaa, a dhaatu must take specific suffixes that change based on lakaara (tense-mood), purusha (first, second, third), and vachana (singular, dual, plural), making the simplest lakaara conjugation table a 3 × 3 = 9 structure. Sanskrit has nava lakaara (nine lakaara), of which shaṭ (six) are tense-based (vartamaana, anadyatana-bhoota, paroksha-bhoota, vibhaashaa-bhoota, saamaanaadhikaraNa-bhavishyat, periphrastic-bhavishyat) and trayaḥ (three) are mood-based (aajnyaa, vidhi, aasheer). This means every dhaatu has a 3 × 3 table for each of these nava lakaara. The nine lakaara referred to above are the ones actually used in grammatical construction. However, when referring to lakaara, dashlakaaraah (the ten mood-tenses) is used. This is because PaNini introduced a tenth lakaara as a catch-all lakaara for handling Vedic verbs that did not conform to the later formalized lakaara rules. By the time the rules of laukika (worldly) Sanskrit crystallized, much of the earlier Sanskrit text—including the Vedas—had already been composed and widely used. These works contained kriyaa-roopas (verb-forms) that did not align with later lakaara rules. This lakaara does not impose rigid paradigms, nor does it have tables like the other nine lakaara, but instead accommodates irregular verb forms that appear in texts, particularly in the Vedas and early BraahmaNas. This special lakaara was instituted to preserve and legitimize ancient verb forms as a linguistic concession to the sacred nature of Vedic texts in deference to the rishis (sages) who created them, freeing them from rules of the laukika Sanskrit allowing them to coexist with the more structured verb system of PaNinian grammar without forcing them into artificial conformity and ensuring their continued existence even as grammatical rules evolved.
Beyond lakaara, two more attributes affect kriyaa-roopa (verb conjugation):
Additionally, kriyaa are categorized as sakarmaka (transitive) or akarmaka (intransitive), where only sakarmaka kriyaa can be used in karmaNi prayoga. If a kriyaa is ubhayapadi and sakarmaka, it will require four different conjugation tables per lakaara (active and passive for both pada).
Sanskrit’s kriyaa system is highly complex, requiring knowledge of nearly dvi-sahasra (2,000) dhaatu, their lakaara classification, pada, karma, and conjugation tables based on these attributes. Furthermore, a dhaatu’s lakaara conjugation pattern is influenced by its dhaatu-gaNa (verb class), of which there are dasha (ten), each governing a distinct set of suffixes—most likely originating from phonetic and semantic variations in historical usage. This intricate system makes Sanskrit kriyaa morphology one of the most challenging yet mentally stimulating aspects of the language for students and scholars alike.
Now, we come to Sanskrit avyayas (whew!). These, thankfully, do not change form. The most important among these are kriyaavisheshaNa (adverbs). Many kriyaavisheshaNa-s derive from naama-pada by attaching -m or -tah suffixes to the praatipadika, similar to English (kind-ly from kind, happi-ly from happy). However, samdarbha (context) needs to be understood before determining the correct meaning. One also needs to watch out for the use of saptamee (locative), panchamee (ablative), and trteeyaa (instrumental) vibhakti forms as surrogates for kriyaavisheshaNa, e.g., shaantau (“in peace”) or shaantena (“with peace”) instead of shaantam (“peacefully”), or poorvaat (“from before”) instead of poorvam (“earlier”).
Finally, Sanskrit has other avyayas, including samyojaka (conjunctions), upasarga (prefixes), neepaata (particles), vaakyaalankaara (interjections, exclamations), and a set of very potent prefixes, which were discussed earlier in the section on lexicon construction. Most of these avyaya function as they do in other languages, but some have uniquely Indian and Sanskrit usages. As with any language, once one develops an intuitive grasp of bhaashaa-pravaaha (flow of the language) and Sanskrit-bhaashaa-samaaja-prashanga (socio-cultural context of Sanskrita language), these determinations become second nature rather than a cerebral exercise.