Implicational hierarchy¶
A linguistic ranking in which the presence of a higher property predicts the presence of lower prerequisites in a defined comparison population.
Core Idea¶
An implicational hierarchy in linguistics is a proposed order of properties: if a case has a higher property, it should also have the lower prerequisites. For a four-step chain A < B < C < D, a case may have A+B+C, but C without B is a gap against an exact version of the proposal. Researchers must first say what one case is, how each property is recognized, and whether the prediction is absolute or a tendency.[1][2]
Scope of Application¶
In syntax, Keenan and Comrie compare primary relative-clause strategies by which noun-phrase functions they can directly relativize.[1] Their Malagasy example is subject-only; a passive route to an object-relative meaning does not turn the primary strategy into direct object relativization.[1] In morphology, number systems can add distinctions such as dual and trial to singular and plural.[2] Malouf and colleagues display one Larike human second-person subject-prefix row with all four distinctions; their English and Upper Sorbian examples are contextual systems in different grammatical domains, not rows of that Larike paradigm.[2]
Clarity¶
The arrow is directional: trial implies dual in the proposed number chain, while dual does not imply trial.[2] A permitted combination need not occur in any language. An apparent exception must be checked against the original coding rule; mixing a language's different relative-clause strategies or Larike prefixes and free pronouns changes the case being tested.[1][2] The number hierarchy is reported as robust, not fully exceptionless.[2]
Manages Complexity¶
Four independent yes/no properties have sixteen possible combinations. A strict four-step chain permits only five lower prefixes. This makes gaps easy to locate, but one binary row can hide optional forms, multiple grammatical strategies or a form whose category is disputed.[1][2]
Abstract Reasoning¶
Choose one comparison unit and code its properties. Check whether the resulting set is a lower prefix. If a true higher property lacks a lower prerequisite, verify the description and then treat it as a counterexample to an absolute rule or an exception to a measured tendency.[2] A paucal category is not automatically a trial and cannot be substituted into a trial⇒dual test.[3]
Knowledge Transfer¶
The same test of coded properties and missing prerequisites can be used for unlike linguistic subjects, including relative-clause syntax and grammatical-number morphology.[1][2] Their feature definitions and evidence do not transfer with the test. Removing the linguistic frame leaves the ranked, asymmetric prerequisite structure of the live Hierarchy Prime; that curated graph link does not transfer linguistic evidence outside the field.
Example¶
Keenan and Comrie contrast a Malagasy subject relative with an unsuccessful direct object-relative construction.[1] Passive promotion can make the object a subject before relativization.[1] Mapped back: the case is a primary relative-clause strategy; the features are directly accessible noun-phrase functions; the coding excludes promotion; subject-only is an allowed lower prefix.[1]
Malouf and colleagues show singular, plural, dual and trial cells in a Larike human second-person subject-prefix paradigm.[2] Mapped back: this row is one fixed case with all four number features; it is compatible with the proposed chain, while a comparably coded true trial without dual would challenge the trial⇒dual step.[2]
Relationships to Other Abstractions¶
Current abstraction Implicational hierarchy Domain-specific
Parents (1) — outgoing typed relations
-
Implicational hierarchy is a decomposition of Hierarchy Prime
A linguistic implicational hierarchy retains ranked levels and asymmetric cross-level prerequisites when its linguistic frame is removed.
Hierarchy paths (4) — routes to 4 parentless roots
- Implicational hierarchy → is a decomposition of Hierarchy → is a kind of Order → presupposes Comparison
- Implicational hierarchy → is a decomposition of Hierarchy → is a kind of Order → presupposes Relation
- Implicational hierarchy → is a decomposition of Hierarchy → is a kind of Order → presupposes Set and Membership
Neighborhood in Abstraction Space¶
Implicational hierarchy sits in a sparse region of the domain-specific corpus (65th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Unclustered & Miscellaneous (2769 abstractions)
Nearest neighbors
- Cue Validity — 0.85
- Cross-Serial Dependencies — 0.84
- Morphosyntactic Alignment — 0.84
- Homoarchy — 0.83
- Differential Evolution — 0.83
Computed from structural-signature embeddings · 2026-10-25
Not to Be Confused With¶
The Noun Phrase Accessibility Hierarchy is one particular syntactic proposal.[1] The Number Hierarchy is a particular morphological proposal.[2] This entry describes their shared linguistic test. Basic color-term development includes a green/yellow branch rather than one total chain.[4] The frozen seed's categorical nasal-inventory claim is not adopted here: Maddieson's survey reports a strong but nonabsolute pattern and a classification-sensitive Irish case.[5]
References¶
[1] Edward L. Keenan and Bernard Comrie (1977), “Noun Phrase Accessibility and Universal Grammar,” Linguistic Inquiry 8(1), 63–99, §§1.2.2 and 1.3.1, pp. 67–70. https://www.jstor.org/stable/4177973 . Original-paper text inspected through a nonpublisher OCR reproduction at https://www.scribd.com/document/346593223/Noun-Phrase-Accessibility-and-Universal-Grammar . registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j
[2] Robert Malouf, Farrell Ackerman and Scott Seyfarth (2015), “Explaining the Number Hierarchy,” p. 1500 and note 1. https://scottseyfarth.com/docs/MaloufAckermanSeyfarth.pdf . Larike data are reported from Laidig (1993) and Siewierska (2013). registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l
[3] Surrey Morphology Group, “Number,” paucal systems. https://www.smg.surrey.ac.uk/features/morphosyntactic/number/ . registry ↩
[4] Paul Kay, “Universality of Color Categorization,” author-hosted paper, PDF pp. 2–5. https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf . registry ↩
[5] Ian Maddieson (1984), Patterns of Sounds, printed pp. 65, 69. https://kcr.digitalwords.net/media/txt/maddieson.i%3A1984%3Asounds.pdf . registry ↩