A morpheme is the smallest unit of language that systematically associates a form with a meaning or grammatical function. In linguistics, morphemes are central to morphology, the study of word structure. The English word cats, for example, contains two morphemes: cat, identifying an animal, and -s, indicating plural number. A morpheme cannot be divided into smaller parts that retain its particular form–meaning relationship. Its form may consist of speech sounds or, in a signed language, visible articulatory features. (ecampusontario.pressbooks.pub)
Morphemes and other linguistic units
A morpheme is not necessarily a word. Some words contain one morpheme, while others combine several. Banana is normally analyzed as a single morpheme despite having multiple syllables; books contains two morphemes despite being pronounced as one syllable. Morpheme boundaries therefore need not coincide with syllable boundaries. Words containing one morpheme are called monomorphemic, and words containing several are polymorphemic. (pressbooks.pub)
Morphemes also differ from phonemes. A phoneme distinguishes linguistic forms but does not necessarily carry meaning itself; a morpheme contributes lexical meaning or information about grammar. For example, the plural ending may consist of a single sound, but its status as a morpheme depends on its grammatical function, not its length. This distinction connects morphology with semantics, which investigates meaning, and phonology, which investigates sound patterns. (ecampusontario.pressbooks.pub)
Free and bound morphemes
A free morpheme can occur independently as a word, as in cat, swim, or new. A bound morpheme cannot occur independently and must combine with other material. English plural -s and past-tense -ed are bound: neither functions as an independent word expressing the relevant grammatical information. Free and bound describe distribution rather than whether a morpheme has a concrete or abstract meaning. (pressbooks.pub)
A root supplies the core of a word and is not further divisible within the analysis being used. An affix attaches to a base, which may itself contain several morphemes. In librarianship, library is the root, librarian is the base to which -ship attaches, and the complete word has a layered structure rather than simply an unordered collection of pieces. (socialsci.libretexts.org)
Affixes are classified partly by position. A prefix precedes its base, as inter- does in international; a suffix follows it, as plural -s does in cats. An infix occurs inside a base, while a circumfix has components on both sides. Some signed-language affixes are articulated simultaneously with their bases rather than sequentially. Consequently, morphemes need not always appear as separate strings placed before or after a root. (socialsci.libretexts.org)
Lexical and grammatical functions
Morphemes can contribute lexical content or grammatical distinctions, although these functions do not map neatly onto the free–bound distinction. Roots often supply lexical content, while affixes frequently modify meaning or mark grammatical features. Two important processes are derivation and inflection. (pressbooks.pub)
Derivation creates a new lexical item. For example, adding -able to wash produces washable, with the meaning “capable of being washed.” This operation also changes the grammatical category from verb to adjective. Derivation does not invariably change category: prefixes can alter a word’s meaning while preserving its category. Compounding, another word-formation process, combines lexical items to form a new one. (vulms.vu.edu.pk)
Inflection instead supplies forms of a lexical item that express grammatical distinctions. English examples include plural books, past-tense walked, and third-person singular runs. Inflection interacts with syntax, since a sentence’s structure may require particular grammatical forms. Languages differ in which distinctions they express through inflection; some use separate words or particles where others use bound morphology. (openlibrary-repo.ecampusontario.ca)
Allomorphs and variation
An allomorph is a variant realization of a morpheme. English regular plurals illustrate this distinction: the ending is pronounced [s] in cats, [z] in dogs, and [ɪz] or [əz] in horses. These realizations express the same plural function, with their distribution determined by the final sound of the base. Spelling alone therefore does not reveal the full morphological pattern. (socialsci.libretexts.org)
Other variation depends on particular lexical items rather than a general sound rule. Children has a special plural ending, while mouse–mice expresses number through an internal vowel change. An unchanged plural such as sheep can be represented in some analyses with a zero morpheme, written ∅: a grammatical contribution without an overt additional form. This is an analytical representation, not a claim that an inaudible sound is pronounced. (socialsci.libretexts.org)
Identifying morphemes
Morpheme identification requires systematic relationships between form and meaning, not merely recurring letters. Analysts compare related words, establish what each proposed component contributes, and distinguish present-day patterns from historical origins. A word that once contained several meaningful components may no longer have that structure for contemporary speakers. Conversely, speakers may recognize productive components without knowing their etymology. Analyses can therefore depend on the language variety and the linguistic knowledge being investigated. (ecampusontario.pressbooks.pub)
In natural language processing, morphological segmentation identifies morphemes computationally. It differs from general tokenization: statistical subword units need not correspond to morphemes. Researchers nevertheless use subword representations as inputs or outputs in systems designed to predict morphological segmentation, keeping the computational representation distinct from the linguistic units being identified. (arxiv.org)