Authority caveat. Nobody on this site claims to know Navajo authoritatively. Every Navajo form is copied from a named published source with its example number; none is invented. Formal rules are hypotheses to be tested against Grant's texts.
SOURCED stated in a published source we opened HYPOTHESIS our formalization or inference GRANT'S CONJECTURE Grant's own claim OPEN needs data or reading
Essay 3 · Deep vs shallow inverse
Deep or shallow? Testing the Navajo inverse against Algonquian
2026-11-11 audit pass 2. Apostrophe fixes only. Navajo glottal stops typed as ' or ’ (for example in prefix names like ʼa- and ʼeʼe-, or words like Awééʼ) now print as ʼ (U+02BC), and a build step no longer turns the closing quote of an English translation inside a Navajo row into a glottal stop (‘…the girls.’, not ‘…the girls.ʼ). No form, gloss or translation wording changed.
2026-11-11 audit pass 4. Copied examples: the essay builder dropped the first Navajo word of every example it copied from a working note (for example Witherspoon’s Hastiin łį́į́ʼ yiztał showed as łį́į́ʼ yiztał, so each gloss sat under the wrong word). The words are restored exactly as the working note prints them; no form was changed. See the pass 4 report.
Grant Golland (research by Grok Bot) · 2026-10-27 · status: HYPOTHESIS (deep by every runnable test; binding OPEN)
Abstract
Algonquian has two inverses: a shallow one, which changes only agreement, and a deep one, which raises the patient over the agent. This essay runs Oxford's diagnostics on the Navajo bi- construction. Every test that can be run on Navajo points to a deep inverse in which the patient really takes the subject position. Navajo has no shallow counterpart at all. The plural prefix da- separates Navajo from Oxford's deep inverse, and the decisive binding test still has no data.
Restated on the unified structure (2026-11-15): the inverse is one extra Move of the patient above the raised agent (H3, H21b), trigger and landing site OPEN. See https://navajo-decoded.grant-golland.workers.dev/unified-structure/#us-map. The essay’s data, judgments and citations are unchanged; it is kept as the record of what was argued then.
1. The question
In every Algonquian direct/inverse pair the site opened, the agent stays the agent and the patient stays the patient. Bruening stresses that the inverse is "not a passive". What changes is which participant controls subject agreement SOURCEDSOURCED.
Oxford separates two inverses SOURCEDSOURCED:
Shallow: 3→1/2. Obligatory, done by Agree alone, with no movement.
Deep: 3′→3. An optional voice in which the patient is raised over the agent and binding is reversed.
Which one is Navajo's bi-?
2. The data
The bi- form is complete with no nouns at all, so the choice is made inside the word:
Willie states the answer directly. The inverse "makes the agent an internal argument … and places an argument with an 'internal' theta role in the external argument position" (p. 100). Sandoval & Jelinek (1989:339) agree that the patient or goal becomes the subject SOURCEDSOURCED.
Every Oxford test that can be run on Navajo points the same way HYPOTHESISHYPOTHESIS: contrast, word order and linking, relativization (standing in for long-distance agreement), and definiteness. On all of them bi- is a deep inverse.
When the patient outranks the agent, bi- is obligatory. That obligatory bi- is the Navajo counterpart of Oxford's forced deep inverse, not of the shallow inverse with first and second persons HYPOTHESISHYPOTHESIS. The site states it as rule RCC, a ban on a higher-ranked slot being c-commanded by a lower one. RCC fits all eight readings of Willie's (72).
Navajo has no shallow side at all: there is no inverse for a third person acting on a first or second person (Willie p. 86) SOURCEDSOURCED.
4. Counter-evidence and differences
The plural prefix. In Oxford's deep inverse, agreement follows the patient. Navajo da- follows the agent in both voices. In that respect Navajo is closer to Bruening's Passamaquoddy analysis, labelled IA-1* on the site, where an EPP feature on little v raises the patient's slot only in the inverse SOURCEDSOURCEDHYPOTHESISHYPOTHESIS.
A possible second hook for da-.Da- may have an optional second hook that pluralizes the patient once the patient has moved into the subject slot. This would be weak extra evidence for movement. It needs da- to sit high, and Hale's last placement puts it high HYPOTHESISHYPOTHESIS.
Rival analyses.
IA-3, Hale 2003's in-situ topic marking, still fits.
IA-2, Hale 1997, has internal tensions but meets no contradicting data.
IA-0, inverse linking, is disfavoured by da-.
See the working notes HYPOTHESISHYPOTHESIS.
5. Status
Binding is the decisive test. No source opened has Navajo binding data for this question. Possessor coreference fits "deep" but does not decide it (Essay 4). Whether RCC is a hard rule, a soft one, or varies between speakers is also open OPENOPEN.
No source mentions chemistry, and no chemistry analog is claimed here.
References
Oxford. Algonquian (Routledge chapter), §§3–4.2. Cited on the site as Oxford 2023. opened
Bruening 2005. Manuscript on the Passamaquoddy inverse. opened
Willie, MaryAnn 1991. Navajo pronouns and obviation (dissertation), pp. 28–32, 72–100. opened
Sandoval, Merton & Eloise Jelinek 1989. The bi-construction and pronominal arguments in Apachean. opened
Hale, Ken 1997 (workshop notes, hale023), 2001, 2003 (manuscript). opened
Bliss 2005 (Blackfoot), cited via Oxford. not opened