Lexeme and Word Form
A **lexeme** is an abstract lexical unit represented by a family of inflected word-forms and associated with one or more conventional lexical senses. Thus **λύω, λύεις, λύομεν, ἔλυσα, λέλυκα** are different **word-forms** belonging to the same lexeme **ΛΥΩ**. A lexeme may be polysemous: dictionary senses are organized under the lexeme rather than requiring a separate lexeme for every contextual meaning.
A **word-form** is a concrete morphological realization of a lexeme in a particular grammatical configuration. An occurrence of a word-form in an actual text is a **token**. Keeping lexeme, lemma, word-form, and token distinct is essential for morphology, lexicography, concordance work, and corpus analysis.
Lemma
A **lemma** is the conventional **citation form** chosen to represent a lexeme in a dictionary, grammar, concordance, or morphological database. The lemma is therefore a lexicographic label, not the lexeme itself.
Greek nouns are conventionally cited by a nominative singular form when available. Greek verbs are normally cited by the **first-person singular present indicative form when available, in the lexically appropriate voice/form**: for example **λύω**, but also middle-form lemmas such as **ἔρχομαι**, **δύναμαι**, and **γίνομαι**. Lexica use conventional alternatives where the expected present citation form is absent, defective, or otherwise unsuitable. Recognizing the lemma allows a reader to connect an encountered word-form with its lexical entry.
Forms ἠγάπησα (aorist), ἀγαπῶμεν (subjunctive), and ἀγαπῶν (participle) all belong to the lemma **ἀγαπάω** 'I love'.
Root
A **root** is the minimal lexical core posited in a morphological or historical analysis, before derivational and inflectional material is added. In transparent families, a root can be seen across related formations; for example **γραφ-** is visible in **γράφω**, **γραφή**, and historically related forms such as **γράμμα**.
Roots are analytical abstractions and are not equally transparent in every lexeme. Sound change, stem formation, and **suppletion** can obscure or replace expected root material. GOLD therefore distinguishes a synchronic morphological root/stem analysis from deeper historical reconstruction rather than forcing all principal parts of an irregular verb into one visibly recoverable root.
Stem
A **stem** is the morphological base used for a particular inflectional formation — the form to which inflectional material is attached. A single lexeme may employ several stems. For **λύω**, for example, GOLD distinguishes formations built on **λυ-**, **λυσ-**, and **λελυκ-** in different parts of the verbal system.
A stem may contain a root together with derivational material, reduplication, tense/aspect formatives, or other morphological elements. The exact segmentation of a form should be justified by the morphology rather than assumed from its surface spelling.
**λόγος** (nom. sg.) and **λόγου** (gen. sg.) belong to the same noun lexeme; their contrast illustrates how a nominal stem participates in a declensional paradigm with different case/number morphology.
Prefix
A **prefix** is a bound morpheme attached before a base or stem. Greek prepositions frequently function as **preverbs** in compound verbs and can alter lexical meaning, aspectual nuance, or argument structure. Examples include **ἐκ-** in **ἐκβάλλω**, **συν-** in **συμβάλλω** (with assimilation of the prefix-final nasal), and **ἐπι-** in **ἐπιβαίνω**.
Morphological boundaries must correspond to real morphemes: **ἐπιβαίνω** is segmented **ἐπι-βαίνω**, not **ἐπ-ι-βαίνω**.
Suffix
A **derivational suffix** is a bound morpheme added to a base or stem to form a new lexeme or lexical class; it normally precedes any final inflectional material. Greek derivation employs many such suffixes. Examples include **-ικ-** in **νομικός** ('legal'), **-μα** in result/action nouns such as **πρᾶγμα**, and **-της** in agent nouns such as **ποιητής**.
Derivational suffixes must be distinguished from **inflectional endings**, which realize grammatical contrasts within the paradigm of an existing lexeme. Derivational morphology is treated systematically in Chapter 69.
Inflectional Ending
An **inflectional ending** is a terminal morphological element that contributes grammatical information without creating a new lexeme. In nominals, endings characteristically contribute **case and number** information; a noun's gender is normally a **lexical property** rather than something newly selected by each case-form.
In finite verbs, personal endings characteristically contribute **person and number** and participate with the stem, tense/aspect or voice formatives, thematic vowel, augment, reduplication, and other morphology in realizing the full verbal feature bundle. Thus in **λύ-ο-μεν**, **-μεν** is the personal ending while **-ο-** is the thematic vowel. It is therefore misleading to attribute tense, mood, voice, person, and number indiscriminately to the ending alone.
Derivational Morphology
**Derivational morphology** forms **new lexemes** from existing bases through prefixes, suffixes, compounding, and related processes. Examples include **γράμμα → γραμματεύς** ('letter/document' → 'scribe') and **ἀπό- + στέλλω → ἀποστέλλω** ('send away'). Derivation may alter lexical meaning, word class, or valency and normally creates a new dictionary entry. (See Chapter 69.)
Inflectional Morphology
**Inflectional morphology** realizes the grammatical forms of a lexeme without creating a new lexeme. Nominal inflection expresses contrasts such as case and number; verbal inflection organizes tense-form/aspect, voice, mood, and, in finite forms, person and number.
Inflection therefore produces members of a paradigm — for example the case-forms of a noun or the conjugated forms of a verb — rather than merely indicating a syntactic 'role'. Much of this grammar is devoted to nominal declension (Chs. 10–14) and the verbal system (Chs. 23–31).
Morphemes and Allomorphs
A **morpheme** is the smallest recurrent morphological unit associated with lexical meaning or grammatical function. An **allomorph** is a contextually conditioned variant realization of the same morpheme.
A clear Greek example is the prefix **συν-**, which appears with assimilated shapes such as **συμ-** before labials and **συγ-** before velars (for example **συμβάλλω**, **συγγράφω**). The variants belong to one morphological prefix but differ in surface form because of regular phonological conditioning. Greek is often **fusional**, so a single ending can realize several grammatical features at once; one should not assume a one-to-one correspondence between every feature and a separate visible morpheme.
Declension
**Declension** is the organization of nominal inflection across **case and number**. Nouns, adjectives, pronouns, articles, and participles participate in declensional paradigms, but they do not behave identically with respect to gender: a noun normally has **lexical gender**, whereas adjectives, articles, many pronouns, and participles inflect for gender in agreement with the noun or construction.
Greek has three major traditional noun-declension classes: the **first**, **second**, and **third** declensions. These classes organize recurring stem and ending patterns (Chs. 10–14).
Conjugation
**Conjugation** is the organization of verbal inflection across the Greek verbal system: **tense-form/aspect, voice, mood**, and, in finite forms, **person and number**. Non-finite verbal forms such as infinitives and participles belong to the verbal system but do not express person; participles additionally inflect like nominals for case, number, and gender.
Greek verbal morphology includes thematic and athematic formations together with important subclasses such as contract and liquid verbs. Conjugation paradigms describe how stems, formatives, thematic vowels, and endings combine (Chs. 23–31).
Agreement
**Agreement** (concord) is the grammatical matching of two or more words. In Greek the chief agreement is **gender-number-case (GNC)** between an article/adjective/participle and its noun (**ὁ ἀγαθὸς λόγος**, **τὴν ἀγαθὴν ὁδόν**). Verbs agree with their subjects in **number and person**. Agreement is the glue of Greek syntax (Ch. 52).
Government
**Government** (also called **case selection** or **rection**) is a lexical-syntactic relation in which a head — such as a verb or preposition — licenses or constrains the form of a dependent complement. In Greek this often includes **case selection**, but it is not always a one-head/one-case rule: a lexeme may participate in more than one construction, and the selected case can interact with lexical sense, complement type, semantic relation, and historical usage.
Prepositions often provide clear beginner examples: **ἐν** selects the dative in New Testament Greek, while **εἰς** selects the accusative. Verbal selection can be more complex. **ἀκούω** occurs with both genitive and accusative complements already in Classical Greek and continues to do so in the New Testament. Classical descriptions often associate the genitive with the person/source or sound perceived and the accusative with the content or thing heard, but actual usage is construction-sensitive and the distinction should not be turned into a rigid rule.
Case selection should therefore be learned as part of a head's broader **valency and constructional behavior**, rather than as an isolated mnemonic such as “this verb takes this case.”
**PT1904 Luke 6:47:** **ἀκούων μου τῶν λόγων** — a genitive construction with **ἀκούω**. **PT1904 John 5:24:** **ὁ τὸν λόγον μου ἀκούων** — accusative **τὸν λόγον** as the content heard. These examples show that **ἀκούω** is not accurately described by a simple “Classical genitive → Koine accusative” rule.
Grammatical Categories
**Grammatical categories** are dimensions of morphosyntactic contrast realized by lexical items and word-forms. In nouns, **case and number** are inflectional categories while **gender is normally lexical**; adjectives, articles, pronouns, and participles can realize gender contrasts morphologically. In the verbal system, relevant categories include **tense-form/aspect, voice, mood** and, for finite forms, **person and number**.
Greek is highly **syncretic**: one surface form can be compatible with more than one feature bundle. Parsing therefore identifies the grammatical values that a form can realize and then uses syntax and context to select among genuinely possible analyses rather than assuming that every form transparently encodes one unique value for every category.
Parts of Speech
Greek words are classified into **parts of speech**: **noun**, **adjective**, **article**, **pronoun**, **verb**, **adverb**, **preposition**, **conjunction**, **particle**, and **interjection**. Some words belong to more than one class depending on use (e.g., **ὅτι** can be a conjunction or an adverb; **πᾶς** can be adjective or pronoun).
Indeclinable Words
**Indeclinable** is a morphological description, not a single part of speech. An **indeclinable nominal** such as the proper name **Ἰωσήφ** can occupy syntactic positions associated with different cases without changing its visible form; its syntactic relation is inferred from constructions such as articles, modifiers, prepositions, governing heads, and the wider clause. By contrast, uninflected **adverbs, conjunctions, and particles** do not bear nominal case at all.
This distinction prevents a category error: absence of overt case inflection on an indeclinable name does not mean absence of a syntactic case relation, while an adverb or particle should not be assigned a nominal case simply because it is also indeclinable. **Ἰησοῦς**, by comparison, is not indeclinable: it has distinct nominative **Ἰησοῦς**, accusative **Ἰησοῦν**, and a syncretic **Ἰησοῦ** used for several other cases.
Clitics
A **clitic** is a phonologically dependent word that forms a prosodic unit with a neighboring word and shows characteristic accentual behavior. Greek distinguishes **enclitics**, which attach to the preceding word, from **proclitics**, which attach to the following word.
Clitics are **not simply words that are incapable of bearing an accent**. Enclitics usually lose their accent in ordinary enclitic position but may be orthotone or receive accent in specified environments; proclitics are normally unaccented but can likewise receive accent under particular conditions. Chapter 5 gives the detailed accent rules and the traditional proclitic/enclitic inventories.
Parsing Methodology
**Parsing** is the disciplined identification of the morphological analysis or analyses licensed by a written form. A robust procedure is: (1) identify the exact form in the text; (2) identify plausible **lemma/lexeme** candidates; (3) segment the form only where the morphology justifies the boundaries; (4) identify stems, formatives, thematic material, and endings; (5) list all grammatically possible feature bundles; and (6) use agreement, government, syntax, discourse, and lexical meaning to disambiguate them.
For a noun, the morphological parse normally reports **lemma, case, number, and lexical gender**. For a finite verb it reports **lemma, person, number, tense-form/aspect, voice, and mood**. A participle additionally carries nominal case/number/gender. **Syntactic function** (subject, object, modifier, complement, etc.) is a separate clause-level analysis and should not be silently folded into the morphological parse. (See Chapter 76.)
Morphological Ambiguity
A single written Greek form can license more than one morphological analysis. The ambiguity must be **formally real**, not merely imagined from English translation. **τιμή** can be nominative or vocative singular; **λύετε** can be present active indicative 2 plural or present active imperative 2 plural.
A particularly useful example is **λύουσι(ν)**: it can be the **3rd plural present active indicative** of **λύω**, but the same surface sequence can also be the **dative plural masculine/neuter present active participle**. Agreement and clause syntax normally distinguish these analyses. Context therefore does not create a parse; it selects among parses that the morphology actually permits.
Syncretism
**Syncretism** is the formal identity of one surface form across two or more distinct grammatical feature bundles. For example, Greek neuter nominative and accusative forms are regularly identical (**τό ἔργον**, **τά ἔργα**), and many verbal endings likewise serve more than one grammatical analysis. Syncretism is one reason context is necessary for parsing.
This descriptive notion should be distinguished from **diachronic functional merger**. The historical absorption of older locative, instrumental, or ablatival functions into later Greek case systems concerns the history and semantics of the case system; it is not identical to the surface-form syncretism of two paradigm cells.
Paradigms
The Anatomy of a Greek Word
A Greek word-form must be analyzed at the correct level: **lexeme** (abstract lexical unit), **lemma** (citation form), **word-form** (concrete realization), and internal morphology such as root, stem, affixes, formatives, and inflectional endings. These categories are related but not interchangeable.
| Term | Definition | Example |
| Lexeme | abstract lexical unit underlying a family of word-forms | ΛΥΩ represented by many inflected forms |
|---|---|---|
| Lemma | conventional citation/headword form | λύω lemma of the lexeme ΛΥΩ |
| Word-form | concrete inflected realization of a lexeme | λύομεν one word-form of ΛΥΩ |
| Root | minimal lexical core posited by analysis | λυ- transparent pedagogical example |
| Stem | base used for an inflectional formation | λυ-, λυσ-, λελυκ- different verbal stems/formations |
| Prefix | bound morpheme before a base/stem | ἀπο- in ἀποστέλλω |
| Derivational suffix | bound element forming a new lexeme/class | -ικ- in νομικός |
| Formative | morphological element contributing to a grammatical stem/form | -σ- in many sigmatic future/aorist formations distinct from the final personal ending |
| Inflectional ending | terminal exponent contributing grammatical features | -μεν in λύομεν 1st plural personal ending |
Chapter glossary
- — lexeme
morphology
An abstract lexical unit represented by a family of word-forms and capable of carrying one or more conventional lexical senses; distinct from lemma, root, stem, and individual word-form.- — lemma
morphology
The conventional citation form used to represent a lexeme. Greek verbs are normally cited by first-person singular present indicative when available, in the lexically appropriate voice/form.- — root
morphology
The minimal lexical core posited in a morphological or historical analysis; not every form preserves the root transparently, especially where sound change or suppletion is involved.- — stem
morphology
A morphological base used in a particular inflectional formation. A single lexeme can employ several stems containing root material plus derivational or grammatical formatives.- — morpheme
morphology
The smallest recurrent morphological unit associated with lexical meaning or grammatical function; Greek morphology is often fusional, so one morpheme can realize multiple features.- — allomorph
morphology
A contextually conditioned surface variant of one morpheme; for example συν-/συμ-/συγ- are phonologically conditioned shapes of the same prefix.- — declension
morphology
The organization of nominal inflection across case and number. Nouns normally have lexical gender; adjectives, articles, many pronouns, and participles additionally inflect for gender.- — conjugation
morphology
The organization of verbal inflection across tense-form/aspect, voice, mood, and, in finite forms, person and number. Non-finite forms do not express person.- — agreement (concord)
syntax
Grammatical matching (esp. gender-number-case) between related words.- — government (rection)
syntax
**Government / case selection** is the lexical-syntactic relation by which a head such as a verb or preposition licenses a complement in a particular case or range of cases. Some heads strongly select one case; others, including **ἀκούω**, participate in multiple constructions whose case pattern interacts with meaning and complement type.- — syncretism
morphology
Formal identity of a surface form across distinct grammatical feature bundles, such as neuter nominative and accusative. This is distinct from diachronic merger of case functions.- — clitic
phonology
A phonologically dependent word that forms a prosodic unit with a neighboring word and exhibits characteristic accentual behavior; includes enclitics and proclitics.