Exploring concepts

The Concepts section offers a semantic route into the corpus. It is useful when a question is about a field of meaning—actions, persons, states, time, communication, substances, and similar broad areas—rather than one known word or text.

What the concepts mean

The site currently uses broad WordNet supersenses to classify mapped lemmas. These are practical semantic labels, not a Sanskrit-native philosophical taxonomy and not a substitute for reading a word in context. A lemma can have more than one label, especially when it has several senses or is used in different ways.

This is meaningful as a discovery tool: it lets you gather a comparison set of vocabulary that has been mapped to the same broad semantic area, then test that grouping against real occurrences. It should not be treated as a definitive translation, a historical etymology, or a claim that Sanskrit concepts divide the world in the same way as English WordNet.

Start with supersenses

The default Supersenses view is the most useful entry point. Each card gives a broad label, a description, and corpus measures:

• Lemmas are the mapped lexical items in the current corpus.
• Occurrences are the combined attested instances of those lemmas.
• Texts are the works in which those mapped lemmas occur.

Large counts describe coverage in this corpus, not importance in Sanskrit thought. Small counts can still lead to highly significant passages.

Open a supersense to see its mapped lemmas. Select a lemma to open its full word entry, then use occurrences and concordance samples to return to actual passages. This passage-first check is the important step: semantic labels are a way to find material, not evidence by themselves.

Use the local neighborhood

On a concept page, the IS-A chain and local graph show how the imported hierarchy relates the current node to broader and narrower labels. Hyponyms are narrower categories below the current node. Click a node or pill to navigate through the local hierarchy.

The graph is intentionally local and bounded. It helps orient you around one concept; it is not a complete visual map of all Sanskrit semantic relations.

Synsets: useful context, limited corpus mapping

The Synsets view exposes the more detailed imported WordNet hierarchy. It can be useful for understanding why a broad category has certain narrower English-language distinctions. However, the current Sanskrit corpus memberships are mapped chiefly at the 28 broad supersense level, not exhaustively to every synset.

For that reason, use a synset page as a reference and navigation aid. Do not infer that an empty or short lemma list means the corresponding Sanskrit idea is absent from the corpus. It may simply not have a direct synset-level mapping yet.

Good research paths

• Begin with a word in the reader, open one of its semantic labels in the word lens, and compare it with other mapped lemmas in that supersense.
• Begin with a supersense such as State or Communication, choose several lemmas, and compare their concordance contexts across literary classes.
• Compare two texts by following the same mapped lemma or a small related set of mapped lemmas, rather than comparing raw totals alone.
• Use a concept page to generate a reading list, then record which passages actually support or complicate the category.

Read the limitations as part of the tool

The concept layer is partial: not every lemma has a semantic mapping, mappings can be broad or ambiguous, and the taxonomy comes through WordNet-based source material. It becomes most reliable when paired with dictionary definitions, morphology, root information where relevant, and the passages themselves.