12. Specification
The Specification is the reference manual (and, effectively, the “grammar”) of TEI Lex-0. It tells you exactly which TEI constructs are in scope and how they are allowed to combine. When you need to answer questions like “Can this element go here?”, “Which attributes are permitted?”, or “What values are valid?”, the Specification is the first place to look.
The Specification is based on the TEI Lex-0 schema: it lists the allowed elements, their attributes, and their content models (what may appear inside them), along with the reusable building blocks that make those content models manageable (model classes, attribute classes, macros, and data types).
You do not need to read the Specification front-to-back. Most encoders will use it as an aid:
- when choosing a tag: look up the element and read its definition and typical usage.
- when structuring entries: check an element’s content model to see what can go inside it.
- when validation fails: use the Specification to understand what the schema expected at that point.
- when mapping or converting data: the Specification provides a stable target for transformation rules.
- when collaborating across projects: it helps you confirm that you are using the same constructs in the same way.
The Specification is also the basis for the TEI Lex-0 schema (RNG, RNC, and XSD). These schemas are meant to be processed by computers and can be used to validate TEI Lex-0 documents for structural and attribute correctness. Do not try to read the schema, unless you suffer from insomnia.
One important note: some recommendations in the narrative Guidelines (for example, ‘use this element only for X’, or ‘if you do A, you should also provide B’) may express semantic or project-policy constraints that are difficult to enforce automatically in a schema. In those cases, the narrative guidance remains essential even when the Specification cannot “force” the rule.
