How words and sentences are built.
10 topics · 19 curated works
No prior grounding assumed.
Locus of Marking in the Clause
Johanna Nichols & Balthasar Bickel · 2005
Maps whether a language marks the grammatical relation between a verb and its arguments on the verb itself, on the dependent noun phrase, on both, or…
+1 more at this level
Assumes you know the vocabulary.
Word Grammar
Richard Hudson · 1998
Presents, in the author's own summary, a theory in which syntax, morphology and meaning are all represented as a single network of labelled…
+2 more at this level
Primary sources and full treatments.
The Philosophy of Grammar
Otto Jespersen · 1924
Argues that grammatical categories like noun or subject should be defined by their function in a sentence rather than by fixed formal criteria, and…
+13 more at this level
12 of 19 works
Johanna Nichols & Balthasar Bickel
Maps whether a language marks the grammatical relation between a verb and its arguments on the verb itself, on the dependent noun phrase, on both, or on neither, and shows this locus-of-marking choice is areally clustered rather than randomly distributed.
link checked 17 Sept 2026Norvin Richards (MIT OpenCourseWare)
Argues that speakers know a rule system they were never taught, and uses phonology, morphology and syntax to show what that unconscious knowledge must contain.
link checked 17 Sept 2026Richard Hudson
Presents, in the author's own summary, a theory in which syntax, morphology and meaning are all represented as a single network of labelled dependency links between words, rather than as separate grammatical components.
Olivier Bonami, Gilles Boyé, Georgette Dal, Hélène Giraudo & Fiammetta Namer
Traces the lexeme concept back to Matthews's distinction between word-form, syntactic word and lexeme, arguing that a word's inflected forms are abstractions over whole word-forms rather than assemblies of morphemes, and surveys how that word-and-paradigm view has developed since.
István Kenesei
Surveys word-class systems from Dionysius Thrax onward and argues that word classes are not a fixed universal inventory but comfortable generalisations that dissolve into clusters of distinguishing properties once tested against a wide range of languages.
Otto Jespersen
Argues that grammatical categories like noun or subject should be defined by their function in a sentence rather than by fixed formal criteria, and that the traditional parts of speech do not hold across languages as cleanly as grammarians assumed.
link failing as of 17 Sept 2026Charles F. Hockett
Compares rival ways of segmenting words into morphemes and shows that no single procedure handles every language's word-formation cleanly, motivating the item-and-process alternative to item-and-arrangement morphology.
link checked 17 Sept 2026Rulon S. Wells
Formalises the procedure for splitting a sentence into nested constituent parts, turning an informal parsing intuition into an explicit method that later phrase-structure grammars built on directly.
Noam Chomsky
Argues that a language's sentences are generated by an explicit formal system of phrase-structure and transformational rules rather than acquired as a list of habits, founding generative grammar as a research programme.
Lucien Tesnière
Proposes that a sentence's structure is a hierarchy of dependency relations radiating from the verb rather than a nested arrangement of phrases, founding dependency grammar as an alternative to constituency-based syntax.
link checked 17 Sept 2026Joseph H. Greenberg
Surveys thirty languages and shows that a language's basic word order predicts a cluster of other structural properties, such as whether adpositions precede or follow their noun, establishing word-order typology as a source of testable universals.
link checked 17 Sept 2026Noam Chomsky
Argues that children's rapid, uniform acquisition of grammar from sparse input is explained only if a universal grammar constrains the space of possible human languages in advance.
link checked 17 Sept 2026