Which Dictionary Should I Use?
Start with MW unless you already know why another dictionary fits your task better. MW is the public default because it is broad, English-facing, well linked in the atlas, and usually the fastest path from a headword to a usable dictionary record.
Trust Block
- Evidence: dictionary metadata, Reader Lookup v1 coverage, public dictionary pages, source links, and documented comparison outputs.
- Limitations: this page chooses a first stop for dictionary evidence. It does not rank dictionaries globally, translate entries, or use corpus frequency.
- Validation: kept consistent with Reader Lookup v1, dictionary pages, and
docs/UC_RD_02_DICTIONARY_CHOOSER.md; checked bynpm run build. - Owner repo:
csl-atlas. - Next use: open MW first, then use the task table below to decide whether a second dictionary should verify, clarify, or challenge the MW result.
Default Path
word -> MW -> source link -> second dictionary if the task needs one
Use MW first for ordinary lookup, translation support, and public reading. Then choose the second dictionary by the job you are doing:
| Your task | Start | Then open | Why the second step matters |
|---|---|---|---|
| I need a quick English meaning. | MW | AP | AP is often shorter and clearer for reading after MW gives breadth. |
| I need the source trail behind a word. | MW | PWG or PWK | PWG/PWK can show dense citations, variants, and nested detail that MW may compress. |
| I need a traditional Sanskrit explanation. | MW | VCP or SKD | VCP/SKD preserve Sanskrit-Sanskrit glossing and authority conventions. |
| I need to check grammar or gender. | MW | AP, VCP, or SKD | Grammar labels differ by convention; compare before trusting one label. |
| I am studying older English dictionary wording. | MW | WIL | WIL helps reveal older English translation choices and Wilson-line inheritance. |
| I am working in a narrow domain. | MW | The specialized dictionary for that domain | Specialized dictionaries are strong inside their scope, weak as general baselines. |
| I need to compare several dictionaries. | Reader Lookup | Lemma dossier | Start with coverage, then inspect exact source-linked records. |
Why Not Say "Use PWG For Lineage"?
For a reader who does not know Sanskrit lexicographic history, "Petersburg lineage" is not an action. The useful reason to open PWG or PWK is practical: they often keep a fuller source apparatus, more nested preverb/derivative structure, and German definitions that preserve distinctions flattened by a later English lookup path.
So the public instruction is:
Use PWG/PWK when you need evidence trail, variants, or structural detail.
That is different from asking the reader to care about lineage before they know what problem they are solving.
What A First Dictionary Does Not Prove
- A missing result is not proof that a word is absent from Sanskrit.
- MW being the default does not make English evidence stronger than German or Sanskrit evidence.
- A low-tag dictionary can still contain rich prose evidence.
- Dictionary evidence is not corpus attestation.
- Machine-derived labels remain machine-derived unless a review page says otherwise.
Next Actions
| If you see this | Do next |
|---|---|
| MW has a clear result and source link. | Use it, then cite the dictionary record rather than the atlas summary. |
MW has only L. or lexicographer-only evidence. |
Treat it as dictionary-tradition evidence and compare PWG/VCP/SKD. |
| MW has many senses. | Open PWG/PWK or the lemma dossier before choosing one sense. |
| MW has no result. | Try Reader Lookup prefix matches, AP, WIL, or a specialized dictionary. |
| VCP/SKD disagree with MW/AP. | Keep both conventions visible; this is not automatically an error. |
Boundary
This page belongs to csl-atlas because every route starts from dictionary
choice, dictionary records, and dictionary source links.
It does not own DCS passage frequency, standards export, GitHub activity, or grammar analysis outside dictionary records.