Greek Glossary · Compendium

Greek Glossary — Guide to the Data Terms

A plain-language field guide to the abbreviations and technical terms behind the Greek glossary: what the source projects (TAGNT, TBESG) are, and what each data field (Strong number, lemma, surface form, book code, token, representative verse, joined statistics) actually means and how to read it.

Source & license: The underlying text and lexicon come from STEPBible-Data, created by Tyndale House Cambridge and curated by STEPBible.org, released under CC BY 4.0.

1. TAGNT

Translators Amalgamated Greek New Testament

What it is. TAGNT is an open, word-by-word Greek New Testament text. It amalgamates the major scholarly editions of the Greek NT into one tagged text where every word carries its lemma, parsing, and Strong number. It was created by Tyndale House, Cambridge and is now curated by STEPBible.org.

How it is used here. TAGNT is the base data for this whole glossary. Every lemma, frequency count, and book distribution you see was produced by parsing the two TAGNT files (TAGNT_Mat-Jhn.txt and TAGNT_Act-Rev.txt) one word at a time.

Source: STEPBible-Data (GitHub) · STEPBible.org · CC BY 4.0.

2. TBESG

Translators Brief lexicon of Extended Strongs for Greek

What it is. TBESG is a concise Greek lexicon (dictionary) keyed to extended Strong's numbers. For each Greek word it gives a short gloss (a quick meaning) and a fuller definition with grammatical notes and example references. It comes from the same STEPBible dataset as TAGNT.

How it is used here. TBESG supplies the meaning side of each entry. It is matched (joined) onto every lemma by Strong number, filling the tbesg.gloss and tbesg.definition fields.

Source: STEPBible-Data (GitHub) · STEPBible.org · CC BY 4.0.

3. Strong's numbers

The join key between text and lexicon

What it is. Strong's numbers are unique identifiers assigned to each distinct biblical word, originally from Strong's Concordance. Greek words use a G prefix, e.g. G3588 for the article ὁ. Extended Strong numbers add letter suffixes to cover words the original list never numbered.

How it is used here. The Strong number is the key that links the two halves of the data: it lets each TAGNT word be joined to its TBESG lexicon entry. If two forms share a Strong number, they share a dictionary meaning.

Example: the lemma ὁ carries Strong G3588, so its TBESG definition is pulled in by matching on G3588.

4. lemma (headword)

The dictionary form you would look up

What it is. A lemma is the base, dictionary form of a word — the form you would find as a headword in a lexicon (e.g. ὁ). Greek is heavily inflected, so one lemma can appear in the text under many different endings; all of them "belong to" the same lemma.

How it is used here. The glossary has one row per lemma. Its freq counts every occurrence that reduces to that lemma, and its meaning is attached once at the lemma level.

5. surface form

The word exactly as it appears in the text

What it is. A surface form is the actual inflected form a word takes in a given verse — the letters as printed — before it is reduced to its lemma. One lemma yields many surface forms depending on case, number, tense, and so on.

Example: in Mat 1:2 the word τὸν is a surface form of the lemma ὁ. Same headword, different ending.

How it is used here. Each lemma's rep_verse.text_el shows one real surface form in its verse context, so you can see how the dictionary form looks in running text.

6. NT book codes (27 books)

Three-letter codes for the New Testament books

What it is. STEPBible/TAGNT identifies each of the 27 New Testament books with a short three-letter code. In canonical order: Mat, Mrk, Luk, Jhn, Act, Rom, 1Co, 2Co, Gal, Eph, Php, Col, 1Th, 2Th, 1Ti, 2Ti, Tit, Phm, Heb, Jas, 1Pe, 2Pe, 1Jn, 2Jn, 3Jn, Jud, Rev.

How it is used here. The book_dist field keys each lemma's frequency by these codes, and verse references such as Mat.1.2 use them as the book part.

7. TAGNT token

One running word in the text

What it is. A token is a single word-occurrence in the TAGNT text — one running word as it appears, counted individually. Tokens are the raw units; grouping tokens by their lemma is what produces headword entries.

How it is used here. The freq figure counts tokens, not lemmas. A frequency of 20,754 for the lemma ὁ means that word occurs as 20,754 separate tokens across the New Testament.

8. representative verse

One fixed example citation per lemma

What it is. The representative verse is the lemma's first occurrence in canonical New Testament order, pinned as its standing example. It records three parts: ref (the reference, e.g. Mat.1.2), text_el (the Greek surface form there), and gloss (a short meaning).

How it is used here. Fixing the first occurrence gives every lemma one concrete, reproducible citation instead of an arbitrary pick, so the example never shifts between runs.

9. TBESG joined (statistics)

How completely the lexicon attaches to the text

What it is. "TBESG joined" describes the result of matching TBESG lexicon entries onto the TAGNT lemma rows by Strong number. The join statistic reports how many lemma rows received a lexicon entry.

How it is used here. In the current build the join is complete: 5,412 / 5,412 lemmas (100%) carry a TBESG entry. Because coverage is total, every row holds both its TAGNT frequency data and its TBESG gloss and definition — no lemma is left without a meaning.