Theta-phenomena tracker (running)
What’s claimed and how well supported
- Labels on this page
- SPECULATIVE ×8 NONE ×1
- Strongest label used
- SPECULATIVE
- Main caveat (quoted)
- - Chemistry relevance has its own label: NONE, ◆ SPECULATIVE, or (never yet used) a sourced analog grade.
Counted automatically from the labels printed on this page. Counts are a reading aid, not a grade of the page.
Terms on this page (1), plain-language definitions
- Valency (ling.)
- Arity: how many actants a predicate takes; formal vs semantic valency can diverge.
Definitions quoted from the Chem for linguists primer glossary.
Started: 2026-10-23 PT (overnight run). Append rows; do not delete. Edit a row only to correct it, and note the date. Purpose: Grant's brief (2026-10-22): "keep track of theta phenomenon for if they pertain to chem stuff". Each row is a grammatical phenomenon that does something with theta roles (assigns, ranks, re-positions, adds or removes them). Each row has its linguistic status and, separately, a chemistry-relevance label. Rules for this file: - Linguistic status is one of SOURCED, HYPOTHESIS, GRANT'S CONJECTURE, OPEN. - Chemistry relevance has its own label: NONE, SPECULATIVE, or (never yet used) a sourced analog grade. - No row is a correlate. Bridge A stays locked and SPECULATIVE. No joint-field claim. - "Key opened sources" lists only sources a run actually opened; see the linked note for page numbers and what was not opened.
Table
| # | Phenomenon (language / family) | What it does with theta roles | Status | Key opened sources | Possible chemistry relevance |
|---|---|---|---|---|---|
| T-1 | Direct / inverse, SAP context (Algonquian: Plains Cree, Ojibwe/Algonquin, Passamaquoddy, Blackfoot) | Nothing to the roles. When a 3rd person acts on 1st/2nd, the patient controls the subject-agreement slot, and the inverse theme sign (-ikw/-igw/-ku/-ok) marks it. Obligatory; no movement ("shallow" inverse: Agree plus impoverishment). | SOURCED (description and analysis: Oxford 2023, 2019); HYPOTHESIS (our "same roles, marked differently" gloss) | Oxford 2023 Syntax 26 preprint; Oxford 2019 NLLT 37 preprint; Oxford, Routledge "Algonquian" ch.; Bruening 2005 ms.; Bliss 2005 WCCFL 24 | SPECULATIVE, specification only. If a chemistry ranking-edge case exists, it should show an obligatory, feature-fixed marking with no rearrangement. Not tested on any chemistry case. |
| T-2 | Direct / inverse, third-person context (same languages; obviative→proximate, or obv→obv by topicality) | Roles fixed and valencyArity: how many actants a predicate takes; formal vs semantic valency can diverge. fixed (not a passive), but the patient is raised over the agent ("deep" inverse = inverse voice). Optional, chosen by topicality. Binding, word order and long-distance agreement follow the patient. This is the best candidate for the Morphic bot's "crossing flip". | SOURCED (Oxford 2023 §4; Bruening 2005); HYPOTHESIS (the "crossing flip" gloss is ours; the phrase is not in any source) | Same as T-1 | SPECULATIVE, specification only. Candidate scouts already on file: Spinel inversion ranking edge, TBP apicophilicity ranking edge, EAS directing ranking edge. New test to run on them: optional, context-chosen rearrangement with roles kept vs. obligatory marking (T-1). Morphic's braid σ/σ⁻¹ reading is their own SPECULATIVE framing. |
| T-3 | Blackfoot sentience ranking and "no non-sentient agents" (Blackfoot, Plains Algonquian) | Third persons are ranked SENT > NON-SENT for direct/inverse, not PROX > OBV. Non-sentient nouns cannot be agents: 'knife' becomes an instrument adjunct (prefix iiht- 'means') with an unspecified subject. | SOURCED (Bliss 2005 (4), (9)/(13), (11)); HYPOTHESIS (that this is ranking by one Dowty proto-agent entailment, sentience) | Bliss 2005 WCCFL 24 pp. 61–66 | NONE found. Relevant to Grant's proto-role thesis, not to chemistry. |
| T-4 | Navajo yi-/bi- inverse (Navajo, Athabaskan) | Roles fixed. In 3/3 clauses an animacy/topicality ranking chooses the yi- (direct) or bi- (inverse) form. Willie: the inverse "always makes the argument with an agent theta role the internal argument." | SOURCED (Willie 1991 pp. 96–97; Willie & Jelinek 2000; Creamer 1974, as read on Navajo Decoded); HYPOTHESIS (that it matches the Algonquian deep inverse, T-2); GRANT'S CONJECTURE (ranking of proto-role weight) | Willie 1991 (box scan; pp. 96–97 re-read 2026-10-23); see Navajo Decoded /inverse/, /inverse-agree/ | SPECULATIVE. This is the original ranking-edge spec (Inverse ranking §6) behind the spinel/TBP/EAS scouts. No correlate. |
| T-5 | Navajo fourth person (ji-/ha-/ho-) | Roles fixed. It takes one participant out of the ranked set, so the direct form survives a clash that would otherwise force bi-. Willie: obviation and deference, while yi-/bi- is voice. | SOURCED (Willie 1991; Willie & Jelinek 2000, as on Navajo Decoded /fourth-person/); HYPOTHESIS ("escape hatch" reading) | See Fourth person vs ranking; Navajo Decoded /fourth-person/, /fourth-person-compared/ | SPECULATIVE. Spec point 6 (Fourth person vs ranking §6): a way to exempt one participant from the ranking. Not yet tested on spinel (Spinel dopant site preference takes it up). |
| T-6 | Incremental theme / distance as measure (English; Navajo motion sentences) | Selects which participant "measures out" the event. Grant: in motion events the distance phrase, not the vehicle, is the incremental theme. | SOURCED for English (Dowty 1991; Krifka 1998 ms.; Hay, Kennedy & Levin 1999, as read on Navajo Decoded /distance-measure/); GRANT'S CONJECTURE (Grant's wording); OPEN for Navajo (one sentence only) | Navajo Decoded /distance-measure/ and its source list | NONE. theta-role-molecule-correlate-scout-v0.md (not published here) searched for an incremental-theme correlate and found none ("Fail for incremental theme"). |
| T-7 | Instrument bee (Navajo) | The instrument's postposition stays bee in both 3/3 sentences found, never yee. So the instrument appears to sit outside the yi-/bi- ranking, while 'for', 'to', 'with (someone)' are ranked. Compare T-3: in Blackfoot, a non-sentient instrument is an adjunct, not an agent. | SOURCED (two NLA sentences, as on Navajo Decoded /instrument-voice/); HYPOTHESIS (outside-the-ranking reading; weak, n = 2) | Navajo Decoded /instrument-voice/; Bliss 2005 (9)/(13) for the Blackfoot side | NONE. |
| T-8 | Valency changing: causatives (Navajo ł-causer vs complex y- + ł + oblique bi-) | Adds a causer role. Which intransitives take the plain ł-causer may follow from the sole participant's entailments (sentient + atelic → complex), or it may be listed by hand. | SOURCED (Hale & Platero 1996, as on Navajo Decoded /causative-split/); HYPOTHESIS (two-entailment rule R2); OPEN (T4 still open) | Navajo Decoded /causative-split/; Causative split | SPECULATIVE, at event-structure grain only. Dopant produce vs maintain: the metals literature separates producing from holding a configuration by stage and form, not by element. Not a correlate. |
Change log
- 2026-10-23 PT (overnight): file started. T-1–T-3 from The Algonquian inverse: same roles, a different subject slot (new). T-4–T-8 summarise existing notes and Navajo Decoded pages, with no new reading except Willie pp. 96–97 (re-read on page images).
Candidates not yet entered (OPEN)
- Possessor raising (Navajo Decoded /possessor-raising/): the possessor becomes the object.
- Psych verbs and pleonastic ho- (Navajo Decoded /pleonastic-psych/): the experiencer sits inside a postposition.
- Mayan Agent Focus, which Oxford 2023's conclusion compares with the inverse (not opened beyond that mention).
- Applicatives and secundative alignment (Algonquian ditransitives treat the recipient like a monotransitive object; Oxford, Routledge ch. §4.2).
Corpus note: theta-phenomena-tracker.md. A running list, started 2026-10-23, of grammatical phenomena that assign, rank, re-position, add or remove theta roles. Each row has a linguistic status (SOURCED / HYPOTHESIS / GRANT'S CONJECTURE / OPEN) and a separate chemistry-relevance label (NONE or SPECULATIVE). Sources are only those the linked notes list as opened. No row is a correlate. No chemical stand-in for a theta role. Bridge A unrevised. Not a joint field.