Three nested slots — a VPISH molecule picture
What’s claimed and how well supported
- Labels on this page
- REAL ×1 SPECULATIVE ×10
- Strongest label used
- REAL
- Main caveat (quoted)
- Grant’s correspondence is HIS and ◆ SPECULATIVE.
Counted automatically from the labels printed on this page. Counts are a reading aid, not a grade of the page.
Terms on this page (9), plain-language definitions
- Dopant
- Intentional dilute impurity meant to tune electronic/optical/catalytic behavior without replacing the host as majority phase.
- Vacancy
- Point defect: a missing atom site in the lattice.
- Valency (ling.)
- Arity: how many actants a predicate takes; formal vs semantic valency can diverge.
- Host (matrix)
- Majority element/phase whose lattice dominates the material.
- Solute
- Intentional minority species dissolved in the host (alloying addition).
- Solid solution
- Solute atoms dissolved into the host lattice (substitutional or interstitial) without a separate precipitate phase.
- Grain boundary
- Interface between crystals in a polycrystal; barriers in Hall–Petch strengthening.
- Dislocation
- Line defect enabling plastic slip at realistic stresses; edge vs screw are the two classic geometries.
- Precipitate
- Second-phase particle that forms (often on aging) and blocks dislocations.
Definitions quoted from the Chem for linguists primer glossary.
Date: 2026-10-04 PT
Status: Corpus note. Same-domain syntax and the materials definitions below are REAL where a source or the drawn analysis states them. Grant’s correspondence is HIS and SPECULATIVE.
Audience: A reader trained in syntax, with less chemistry.
Does not: announce a joint field, invent Navajo forms, glosses, or trees, invent concentrations, barriers, thicknesses, or other numbers, invent a chemical stand-in for a theta role, revise Bridge A, or merge this page with the movement, lexicalization, causative, or Zipf pages.
Read against, and not contradicting: Derivational movement; Lexicalization; Causative split; Zipf.
Abstract. Grant is looking for a link between dopantIntentional dilute impurity meant to tune electronic/optical/catalytic behavior without replacing the host as majority phase. structure and theta-role structure. The arrangement he names has three slots. Under syntax names they are the verb phrase, the tense phrase, and the clause. Under his molecule names, once, they are a smaller molecule, a larger molecule, and a still larger one. The slots are the same three. He does not know whether chemistry works like that. No paper states the correspondence. This page does not pretend one has.
The comparative result is narrow. Ordinary English, as one teaching handout draws it, is the clear VPISH hop: a lexical subject starts in the verb phrase and moves to the specifier of the tense phrase, and it is still in the clause. Only that case is given a chemistry step, and only as HIS SPECULATIVE pedagogy. The mechanisms available for that step are three, and only three: an interstitial hop, a vacancyPoint defect: a missing atom site in the lattice. hop, and segregation that stays in the solid. Navajo, on the drawn analysis, leaves arg1 in the verb phrase and spells the subject out as an exponent on mode. Belfast English, as the same handout reports it, leaves the logical subject in the verb phrase. Welsh, only as that handout reports an argument, leaves the subject in the verb phrase and moves the verb. Those three are not given a diffusion step and not given a segregation step. Nesting can still be said: a site inside a larger piece of the same thing, and the participant has not left.
Navajo is the extreme edge case, not a failed English hop. The unit he takes it to be showing is the theta role rather than the lexical item. Chemistry, as opened here, tracks pieces of matter. It has no stand-in for a theta role. None is invented. The gap is kept open. It is not treated as a refutation.
Bridge A, concentration windows set beside valencyArity: how many actants a predicate takes; formal vs semantic valency can diverge. and saturation, is a different speculative strand. See Solubility windows. This page does not revise it.
- 1. How the labels work
- 2. His three levels, in both vocabularies
- 3. What the syntax names mean, before any language is special
- 4. Navajo at the verb phrase
- 5. Navajo at the tense phrase
- 6. Navajo at the clause
- 7. Names that have to stay available
- 8. Three mechanisms, and no others
- 9. The shape that is being tested
- 10. Where the subject really moves
- 11. Where the map is weak, including for English
- 12. The open gap: a theta role is not a piece of matter
- 13. What stays separate
- 14. What this page did not add
1. How the labels work
Three labels are kept apart so a sentence cannot drift from one column into another.
Sourced, same-domain. A claim a named source states, or a claim the movement page already drew from sources it opened. On this page those claims are the Navajo placement of arg1, the handout’s report of where English, Belfast English, and Welsh put the subject, and the plain-language definitions of an interstitial hop, a vacancy hop, and segregation that stays in the solid. They are not a theory of theta roles as chemistry.
HIS. Grant’s picture: the three slots, the molecule names, the balloon, the clown-car cab, the guess that the units which matter are theta roles rather than lexical items, and the guess that a local rearrangement inside one solid is the shape to look at. His sentences are not upgraded because a sourced sentence sits nearby.
SPECULATIVE. Every cross-domain sentence. The whole correspondence is a pattern match. “The sources do not say this” does not drop the question. It is why his sentences and the sourced sentences are kept apart.
A fourth discipline is comparative. A diffusion step or a segregation step is drawn only where the syntax really moves a subject out of the verb phrase. Where the subject stays in the verb phrase, nesting can be said. Diffusion and segregation are not said. Navajo is not described as a hop that failed. It is the edge case that shows a theta role without a lexical item climbing.
This page points at four neighbors and does not absorb them.
Derivational movement asks where arg1 is. It draws the pronominal-argument derivation and records the flagged Spec,M overlay the opened essay rejects. Its own chemistry half is a different pedagogical map. Leaching and evaporation are the wrong parallel on that page. This page does not redraw those trees and does not fold that map into this one.
Lexicalization asks what is listed. The listed unit there is the verb base. The verb theme is the looser non-compositional prefix plus base. That is not this page’s question.
Causative split asks how a short transitivizer, a longer morphological causative, and an analytic causative that was asserted and not exemplified line up. Some ł is thematic. That scout is not a tree, and it is not this picture.
Zipf is ranked frequency. Williams et al. 2015 are cited there. The phrasal-verb question and the quark / anyon analogy stay open on that page. This page repeats only the cut it needs, and it does not choose quarks over anyons.
The typed-participant map asks which participant types a mechanism family can wear. Different question. Not merged.
2. His three levels, in both vocabularies
HIS. Look for a link between dopant structure and theta-role structure. The arrangement to look for is the shape of a Navajo phrasal verb, or of Navajo tense, or of a clause-sized phrase.
HIS, molecule language once. The subject argument starts in the scope of a smaller molecule, the verb phrase. It may move to a position in a larger, superordinate molecule, the tense phrase. It may move again into a still larger one, the clause.
Those are the same three slots under the syntax names: verb phrase, tense phrase, clause. He does not know whether chemistry works like that. This note does not pretend an opened paper has settled it.
| Slot | Syntax name | His molecule name, once | What the slot is for, in his picture |
|---|---|---|---|
| Smaller | Verb phrase | Smaller molecule | Where the subject starts |
| Larger | Tense phrase | Larger, superordinate molecule | Where the subject may move, still inside a bigger piece |
| Still larger | Clause | Still larger molecule | Where the subject may move again |
The table is his arrangement, not a finding. “May” is his wording. Later sections say, language by language, whether a movement of that kind is actually in the syntax. Possibility in his picture is not evidence that a given language performs the move. Ordinary English, as the handout below draws it, has one move, into the tense phrase. It does not then move the subject again into the clause as a third landing. His further move stays his, and it is not drawn.
HIS, optional metaphor only. The hot-air balloon (Major Marvy) is not a physics fact. The bit he now means is local levity: a local lack of the usual pull. It is a way to picture a region where the usual pull is weakened. It is not a claim that gravity fails there. It is not an interstitial hop, a vacancy hop, or segregation. It is not lower free energy at a boundary, and a vacancy is not a balloon. Clown-car cab is his gloss for laughable expandable clauses. It names, for him, the way a clause can grow and still be one clause. It is not a syntactic category in the sources, and it is not a materials mechanism.
HIS, the edge. Navajo is the extreme edge case that shows the inner workings. The units that move, on his view, are theta roles, not lexical items. He says chemistry has no super clear correlate for that. That gap is recorded below. It is not treated as a refutation, and no chemical stand-in is supplied.
3. What the syntax names mean, before any language is special
The verb phrase, in the VP-internal subject hypothesis, is the smaller constituent in which the verb and its arguments are first put together. The external argument — the subject theta-role bearer — starts there. On the usual name for that position, it is the specifier of the voice head v: Spec,vP. “Starts there” means the argument is merged there. Merge is the step that builds the position. It is not yet a claim that the argument later leaves.
The tense phrase is the next constituent up in the familiar English-style clause: a phrase headed by tense, whose specifier is the surface subject position, Spec,TP. Movement, if the language’s analysis has it, is a later step that takes the subject from the verb-phrase position to that specifier. The verb phrase is still inside the tense phrase. The subject has changed site. It has not left the clause.
The clause is the still larger constituent. In the ordinary English case below, it contains the tense phrase, which contains the verb phrase. The subject that has moved to Spec,TP is already inside the clause. The clause is the container of both positions. It is not, in that derivation, a third site the subject then moves into.
arg1, as the movement page fixes the term and as this page reuses it, means the external argument: the subject theta-role bearer, merged inside the verb phrase, not a separate lexical noun phrase. The shared name Spec,vP does not mean every language then moves that argument.
Spell-out is the morphological realization of material that is already in the structure. It can put a subject exponent together with a mode morpheme inside one verb word. Spell-out is not, by itself, a climb to a higher specifier.
None of those definitions is a chemical claim.
4. Navajo at the verb phrase
Sourced. On the drawn pronominal-argument analysis, arg1 — the external argument, the subject theta bearer — is merged in Spec,vP and is not moved. It is realized as the subject exponent inside the verb word. Hale’s opened essay licenses those arguments in place. What puts the subject with mode is spell-out, not a climb out of the verb phrase.
The movement page is where that derivation is drawn, example by example. This page does not redraw the brackets and does not add a form. Hale’s essay, as that page reports the opening, treats person–number morphology inside the verb word as the arguments themselves. An independent noun, when one is present, is a contrastive or discourse adjunct on that drawing. Surface order that does not match the merge tree is, on his proposal, phonological: spell-out into the verb’s skeleton.
The movement page also records a flagged overlay in which arg1 would move to Spec of Mode, because Mode is the label Hale lines up with Infl/T. The essay that supplies the merge sites rejects that overlay. This page does not put it back. A rejected landing site is not a movement the syntax has, and it is not a reason to draw a diffusion step or a segregation step for Navajo.
The verb-phrase slot, for Navajo, is occupied at the start and still occupied at the end of the drawn derivation. The subject theta bearer has a site. The site is inside the verb word.
5. Navajo at the tense phrase
The middle slot is real as a descriptive object. It is not a tense phrase the subject moves into.
Sourced. McDonough says there is no tense in Navajo. Mode is what is inflected for person and number, and subject marking is an exponent on mode. Mode and the subject exponent are fused. The movement page’s descriptive core, from Young & Morgan via McDonough, is the same fusion from the template side: the core verb is mode inflected for person and number, as one paradigmatic piece, plus classifier and stem. This page does not add a page number the picture did not give. The sentence it records is the one above: there is no tense.
Sourced. Smith, Perkins, and Fernald say Navajo is not fully tensed. Every verb word has one mode. Future is the only mode whose main job is temporal location, and even future is not a full tense. Their high “Navajo Infl” is a landing site they assume for the verb head, not for arg1. This page does not re-open that paper and does not supply a bibliographic line the picture did not give. The constraint is enough: the middle position people reach for when they say “tense” is, in these sources, mode, inside the verb word. The verb head may be what those authors land in their high Infl. arg1 is not.
Hale’s remark that mode corresponds to Infl and T is a correspondence between heads, so a reader of English-style trees can see which Navajo head sits in that neighborhood. It is not movement of arg1. McDonough’s “no tense,” and Smith, Perkins, and Fernald’s “not fully tensed,” say what that neighborhood contains.
6. Navajo at the clause
Sourced. A well-formed Navajo clause may be only the verb word. Nominal arguments are optional. In the opened polar questions, a high particle and/or an enclitic is added, and arg1 is not extracted. In the clearest opened content question, the question word stays in the object’s nominal position and the verb-internal person marking stays on the verb. On the drawn analysis a full noun phrase is an adjunct, not arg1 that has climbed out. The clause is the larger piece the verb word already is, or sits in. It is not a third molecule the subject is shown moving into.
The movement page names the polar particle and the enclitic, and it prints the content-question sentence it could read. This page does not repeat those strings as if it had re-opened them. Question morphology can be added at the edge of the clause. The subject exponent does not leave the verb in order to build that question. A full noun phrase, on the drawn analysis, is an adjunct linked to an exponent, not the exponent itself promoted into a higher specifier.
The negatives are part of the pattern. They are not a reason to stop looking at the slots. Each slot is attested as a size of Navajo structure. None of the three is attested as a step in a climb of arg1.
7. Names that have to stay available
Sourced. There is no single Athabaskanist object called the Navajo phrasal verb. Hale’s preverb and the verb stem are one lexical constituent inside one verb word. He compares that discontinuity to a Germanic particle verb and does not call it a phrasal verb. Chee and Yazzie use “phrasal verb” for a different object, a postposition plus a verb word.
His picture asked after “the shape of a Navajo phrasal verb” as one arrangement to look at. The sourced answer is that the name does not pick out one object. The lexicalization page is where listing is argued: verb base versus verb theme, beside his coinage “phrasisized.” This page does not redo that listing.
Williams and coauthors’ “phrases” are a third object. They are random mixed-length contiguous spans inside punctuation-bounded clauses. That cut neither selects nor excludes phrasal verbs. The Zipf page is where the paper is opened: Williams, Lessard, Desu, Clark, Bagrow, Danforth, and Dodds, Scientific Reports 5, 12209 (2015; srep12209). Their phrase is a random partition, not a verb plus its arguments. The further unit in this note is not “Williams’s phrase = the crystal.”
HIS, recorded on the Zipf page and repeated here only as a boundary. Molecule, atom, and quark are his. Williams’s cut, which is theirs, is morpheme as the atom of language and a mixture of words and phrases as the molecule. That is one level off his word-as-atom cut. Their paper does not measure chemical molecules. This page does not choose quarks over anyons. The Zipf page records both as his open analogy and does not choose either. No confinement claim and no particle-physics result is stated here.
The phrasal-verb Zipf question stays open. On the Zipf page it is his: maybe the power law applies to phrasal verbs, a verb plus its arguments, rather than to phrases in general. He does not know if that is true. This page does not test it and does not state an exponent.
8. Three mechanisms, and no others
The definitions in this section are same-domain materials vocabulary. They were opened from DoITPoMS teaching pages and from the Wikipedia digest “Segregation (materials science).” The movement page had already used the mechanisms page and the segregation page. A later pass reopened those and also opened the temperature page, the summary, the microstructural-effects page, and the grain-boundary animation. No equation is copied. No concentration, barrier, thickness, or enrichment ratio is stated.
These are the only chemistry mechanisms this page uses. Anything else in the opened pages — evaporation, a second phase, precipitation, a direct swap, a ring of atoms trading places — is named so it is not used.
A hostMajority element/phase whose lattice dominates the material., in this glossary, is the piece of solid the foreign atom sits in. A site is a place in that piece: a lattice point the host atoms occupy, or a gap between those points. “Dopant,” in his likeness, is his word for the foreign atom. The vocabulary note (Doping vs alloying vocabulary) reserves “dopant” for dilute electronic intent and prefers “soluteIntentional minority species dissolved in the host (alloying addition).” for many structural additions. This page does not reclassify the likeness as Family 1 electronic doping.
8.1 Interstitial hop
Sourced. DoITPoMS, “Diffusion mechanisms.” A small impurity sits in a gap between lattice sites, not on a lattice site, and may step into a neighboring gap if that gap is empty. It does not need a missing atom next door. It does not leave the crystal.
In ordinary words: the atom changes which gap it is in. The crystal around it is the same crystal. No vacancy has to be created first.
8.2 Vacancy hop
Sourced. Same page. An atom that occupies a normal lattice site steps only if a neighboring site is empty. It moves into that vacancy, and the empty site is then the one it left. Direct swapping of two atoms, and a ring of several atoms trading places at once, are not the mechanism in practice. Both would cost too much shoving, and the ring is too improbable. The atom remains in the solid. Needing a vacancy is why this path is slower than the interstitial one. No rate is stated.
DoITPoMS also says it is often useful to think of this second mechanism as the empty site moving, rather than as the atom moving. That remark is a pressure on the analogy, kept for the weakness list below. It is not a cleverer mapping. The atom, either way, remains in the solid.
8.3 Segregation that stays in the solid
Sourced. Wikipedia, “Segregation (materials science).” In this sense, segregation is enrichment of atoms, ions, or molecules at a small region — in a polycrystal, at a dislocationLine defect enabling plastic slip at realistic stresses; edge vs screw are the two classic geometries., a grain boundaryInterface between crystals in a polycrystal; barriers in Hall–Petch strengthening., a stacking fault, or the interface between two phases — by partitioning out of a solid solutionSolute atoms dissolved into the host lattice (substitutional or interstitial) without a separate precipitate phase.. The solute is still in the material. The page uses “adsorption” for the same kind of gathering when the source is a liquid or a gas. It separates this molecular-level segregation from phase separation and from precipitation, which put material into a macroscopic region of a different composition, and from the unrelated use of the word for particles sorting in a granular pile.
A crystal, for a syntax reader, is one continuous lattice. A grain is one such region inside a solid made of many crystals. The boundary between two grains is still inside the solid. A surface may be the outer face of the piece. The boundary is the privileged-site candidate that stays clearly inside. The free surface is the weaker candidate, because it is also where evaporation starts.
Two ways that same gathering can happen, both still inside. Sourced, same page. They are not extra mechanisms. Equilibrium segregation is the arrangement a settled piece prefers: sites at a boundary or a surface have a different energy from sites inside the lattice, and solute gathers there so the whole piece has lower free energy. That preference lessens as temperature goes up. Non-equilibrium segregation is left over from processing. Solute can ride with vacancies that are moving toward a boundary while the piece is quenched or stressed, or it can pile up at a moving interface. That extra enrichment grows as temperature goes up, and it would disappear if the piece were held long enough to even out, because the even distribution is then the lower-energy state. McLean (1957) is the source the page cites for the equilibrium partition between boundary and lattice. Westbrook (1964) is the source it cites for the non-equilibrium idea. Neither book was opened. Their equations are not copied.
A grain boundary as a path, still the same solid. Sourced. DoITPoMS grain-boundary animation, summary, and “Microstructural effects.” Atoms move faster along a grain boundary because the lattice is disrupted there and empty sites are more plentiful. Dislocations are the same kind of shortcut, for the same reason. The boundary is thin, so the shortcut matters when travel along it is much faster than travel through the grain interior, which that page locates at low temperature. The boundary is still part of the solid. The measured diffusivity they write is the interior contribution plus a boundary contribution. No thickness and no diffusivity are copied here.
Heat, still without a number. Sourced. DoITPoMS, “Temperature Effects,” restated on the summary page. A hop of either kind has to get over an energy barrier, and a hotter piece makes a successful hop more likely. The vacancy path has a second temperature dependence, because how many empty sites exist also changes with heat. An interstitial does not need an empty lattice site to be created first, so DoITPoMS says its barrier is much lower and its diffusion much faster. That comparison is theirs. It is not a measured value, and it is not a Navajo fact. Mode is not temperature. This page does not pick a temperature story and call it a linguistic clock.
The exit, named so it is not used. Sourced, same Wikipedia page. At high temperature, atoms can evaporate from a surface, and then the McLean description no longer holds. Evaporation is the case that leaves. It is the wrong parallel for a participant that stays in the clause. A second phase, or precipitation, is the other wrong parallel: the foreign material has left the host solution, even if it is still inside the component as a separate region.
Casting segregation, only as a warning. Sourced, same page, reporting Campbell. All metal castings segregate to some extent. Microsegregation is a composition difference between nearby dendrite arms, over distances small enough that a homogenizing heat treatment can let diffusion even it out. Macrosegregation is a composition difference across the casting that heat treatment cannot undo, because the distances are too large. Both are still in the solid. Macrosegregation is not recruited as “the clause.” “Still in the component” is not the same as “still in the neighborhood.”
The movement page already points at hydrogen in titanium as the corpus’s exhibit of an interstitial, and at carbon in iron as an interstitial alloying addition. This page does not add numbers to those exhibits.
9. The shape that is being tested
HIS. SPECULATIVE. Not a mechanism, not a number, not a joint field. The shape to test is nesting: a birth site inside a small unit, a privileged site that is still inside a larger unit, and a further unit that contains both, with the participant not leaving. Chemistry can show that shape for an atom. It does not show it for a theta role. That difference is the open gap below. It is not a reason to erase the shape.
Birth site, inside the small unit. On the syntax side this is the verb phrase: Spec,vP, where the external argument is merged. On the chemistry side the small unit is the host lattice, and the birth site is either a normal lattice site (the atom has been substituted for a host atom) or a gap between lattice sites (the atom is interstitial). DoITPoMS starts from exactly that picture: an impurity in an otherwise perfect structure sits in one of those two places, and the two places are what make the two diffusion mechanisms. The likeness is only “a place inside the small repeating unit, or inside the verb phrase.” It is not a claim that a theta role is an atom, or that a verb phrase is a unit cell.
Privileged site, still inside a larger unit. On his wording the larger unit is the tense phrase, and the privileged site is its specifier. That wording is safe for ordinary English, and only there, as the next section says. It is not safe for Navajo, for the Belfast associate, or for Welsh as reported. A grain boundary, or a surface, is the chemistry candidate for a privileged site that is not the birth site and is still inside the same piece. The surface is the weaker of the two, because it is also where evaporation starts. The boundary is the one that stays clearly inside. Even for English, a grain boundary is a looser picture than a hop to a neighboring site, for a reason the comparison states: Spec,TP in the handout is not optional and is not a defect.
A further unit, and the participant does not leave. On the syntax side the further unit is the clause. A Navajo clause may already be the verb word. An English clause, in the ordinary case, contains the tense phrase the subject has moved into, and the subject is still in the clause. On the chemistry side the further unit is the crystal, or the component. Interstitial hops, vacancy hops, and segregation to a boundary are rearrangements inside that piece. Evaporation, leaching, and the forming of a separate phase are the rearrangements that leave the host solution. They are not drawn for any language on this page.
Local levity stays on his side of the line. A grain boundary is a place of different energy, and equilibrium segregation is solute gathering there so the free energy of the whole piece is lower. That is a thermodynamic preference. It is not levity, and it is not his balloon. A vacancy is a missing atom that lets a neighbor step. It is not a lack of pull, and it is not an argument position.
10. Where the subject really moves
SPECULATIVE wherever a chemistry sentence is set beside a syntax sentence in this section. The rule is not speculative: a diffusion step or a segregation step is drawn only where the syntax really moves a subject out of the verb phrase. Where the subject stays in the verb phrase, the hop is not drawn.
The source for the comparisons is one teaching handout opened for this picture: Ethan Poole, “VP-internal subjects,” Ling 200B, 17 November 2021, https://ethanpoole.com/handouts/ling-200b/2021/13-vpish.pdf. It states the hypothesis as: subjects begin in the verb phrase and may move to Spec,TP. The “may” is Poole’s. The evidence he walks through is, as he says, from Koopman and Sportiche 1991. That paper was not opened. Woolford 1991, which he cites for verb-initial languages, was not opened. No tree is added here beyond the prose. No example sentence is invented for a variety this page did not quote.
10.1 Ordinary English: the clear hop
Sourced, as the handout draws common English. The subject starts inside the verb phrase and moves to Spec,TP. The EPP on T is satisfied by that move. The object is not what moves, because T targets the closest noun phrase. Auxiliaries can sit between the subject and the verb, including inside idioms whose pieces otherwise want to be one constituent. The handout’s answer is that the idiom is a verb phrase and the subject has raised. A subject can also be read above or below negation, which the handout treats as a choice of which copy to interpret. For this page the geometric point is enough: there are two positions, the birth site in the verb phrase and Spec,TP, and the lexical subject occupies the higher one in the derivation the handout draws, while a copy remains below.
The clause is the unit that contains both positions. It is not a third landing site the subject then moves into. His three-slot wording had a further move into the clause. Ordinary English, as this handout draws it, does not make that third move. A later move of the subject into the complementizer layer was not opened and is not drawn.
HIS. SPECULATIVE. This case only. The cautious chemistry likeness is an atom that starts on a site in the host and relocates to another site in the same solid: a step into a neighboring gap (the interstitial hop), or a step into an adjacent vacancy (the vacancy hop). The subject has not left the clause. Evaporation is still the wrong parallel. Segregation to a grain boundary is a looser second picture, because Spec,TP in this derivation is not optional and is not a defect. Non-equilibrium segregation, solute riding vacancies toward a boundary during a quench or under stress, is a real in-solid relocation, and it imports a processing history the syntax does not have. It is not used as the English picture. The picture that is used is the hop to a neighboring site, still inside the clause’s correspondent, the solid. Even that picture is weak. The weaknesses are the next section. One of them belongs here: a syntactic copy is not a random walk, a grain boundary is not tense, and the low copy can still be interpreted, whereas an atom that has hopped is not, on these pages, still sitting at the birth site.
English shows a lexical noun phrase moving, and the theta role travels with that word. It does not isolate a theta role from the lexical item. The likeness just drawn is a likeness for that lexical movement. It is not an account of a theta role as a unit of its own.
10.2 Belfast English: no hop
Sourced, same handout. Do not draw the hop. The handout’s Belfast sentences put an expletive there in Spec,TP and leave the logical subject inside the verb phrase. In the polar questions, the auxiliary moves past there, which is Poole’s argument that there is in Spec,TP and the logical subject is still down in the verb phrase. The EPP is satisfied by merging the expletive, not by moving the associate. The associate has a birth site inside the verb phrase and does not leave it.
That is nesting. It is not interstitial diffusion, not a vacancy hop, and not segregation. This page does not invent a Belfast sentence to illustrate the point. The handout’s configuration is the constraint: the logical subject stays in the verb phrase, so it is not given a diffusion or segregation step.
10.3 Welsh, as reported: no hop
Sourced only as the handout reports an argument. Do not draw the hop. The handout says it has been argued that verb-initial languages are derived by leaving the subject in the verb phrase and moving the verb to T, and it points to Woolford 1991 for an overview. Woolford was not opened. This page does not adopt a Welsh analysis of its own and does not repeat a Welsh example. What is usable is the caution: a language can show the subject born inside the verb phrase without the subject moving. If that argument is right, the thing that moves is the verb, not the subject.
A chemistry hop of the subject-like participant is not drawn. Moving the verb is not given a dopant story either. No page opened here makes a host atom into a verb.
10.4 Navajo: no hop, and not a failed one
Sourced, drawn analysis. Do not draw the hop. arg1 is merged in Spec,vP and is not moved. The audible subject is the exponent on mode, inside the verb word. There is no tense phrase the subject climbs into. McDonough: no tense. Smith, Perkins, and Fernald: not fully tensed. The clause need not contain a separate lexical subject at all.
The honest likeness is nesting: a site inside a larger piece of the same thing, and the participant has not left. Calling that spell-out a hop, or calling the mode exponent a grain boundary, would force a movement the syntax note refused. It is not done. Navajo is not given interstitial diffusion, not given a vacancy hop, and not given segregation.
HIS. This is the extreme edge case that shows the inner workings. It is not a failed English-style hop. Belfast’s associate and Welsh-as-reported also stay in the verb phrase, and they too are refused a hop. Navajo is the edge among them because the piece you can hear as the subject need not be a separate lexical item at all. On the drawn analysis the argument is the exponent. A full noun phrase, when there is one, is an adjunct. Nominals are optional. A well-formed clause may be only the verb word. English, above, is the opposite exhibition: a lexical noun phrase does move, and the theta role and the word travel together. Navajo lets the role be seen without that word moving, and often without that word being there.
His sentence, that the unit of interest is the role rather than the lexical item, is a description of what the edge case makes visible. It is not a new derivation. This page does not redraw the tree so that a theta role moves while arg1 stays. The sourced sentence remains: arg1 is merged in Spec,vP and is not moved. The refused Spec,Mode overlay is not put back in because the role is interesting.
10.5 What was not surveyed
French, Italian postverbal subjects, and Icelandic transitive expletives are often discussed in this literature. They were not opened. They are not given a movement diagram and not given a chemistry hop. English is the clear case that was opened. The others wait. A later language is copied from this rule, not from a guess: if its analysis moves the subject from the verb phrase to the tense phrase, the English-side likeness is the one that could be repeated, still speculative. If the subject stays in the verb phrase, nesting only, and no diffusion or segregation step.
| Ordinary English | Belfast English, as the handout | Welsh, as reported | Navajo, drawn analysis | |
|---|---|---|---|---|
| Birth site | Verb phrase | Verb phrase, for the logical subject | Verb phrase, if the reported argument is right | Spec,vP |
| What occupies Spec,TP | The lexical subject, after movement | An expletive there, merged there | Not the subject, on the reported argument. The verb is what moves to T | Not drawn. There is no tense phrase the subject moves into. Mode is fused with the subject exponent |
| Subject leaves the verb phrase? | Yes | No | No | No |
| Diffusion or segregation step | Yes, cautiously: interstitial hop or vacancy hop. Grain-boundary segregation is a looser picture and is not the one used as identity. SPECULATIVE | No | No | No |
| What may still be said | The subject is still in the clause | Nesting. The associate stays inside the verb phrase | Nesting, if that argument is right. The verb’s movement is not a dopant story | Nesting. Subject exponent on mode, inside the verb word |
| How to read it | The clear hop | Not a hop | Not a hop | The edge case. Not a failed English hop |
11. Where the map is weak, including for English
These are reasons the likeness can fail. They are not a list of repairs. The map is weak even for ordinary English.
No paper states the correspondence. The match is a pattern, read off pages that were not written to be read together.
Diffusion and segregation move pieces of matter. The drawn Navajo analysis does not move arg1. Belfast’s associate does not move. Welsh, as reported, does not move the subject. Using a hop to describe any of those three puts a step into a derivation that leaves the subject in the verb phrase. The hop is reserved for ordinary English. Nesting is what survives elsewhere.
The middle slot in Navajo is mode, fused with the subject exponent, not tense. Mapping Spec,TP onto a grain boundary imports a landing site the opened Navajo analysis does not give arg1. McDonough’s “no tense,” and Smith, Perkins, and Fernald’s “not fully tensed,” stay in force.
Even in English, the “privileged site” and a grain boundary are different kinds of privilege. Spec,TP in the handout is a position the derivation fills because of the EPP. A grain boundary is a place some solute prefers, or is carried to, while most of the solute can remain in the grain. Optionality on one side and obligatoriness on the other are not the same relation. That is why segregation is only a looser picture for English, and why the cautious likeness actually drawn there is a step into a neighboring gap or an adjacent vacancy.
English, as drawn, has two positions plus a container. His picture has three landings. The clause is not shown as a third site the subject moves into. Forcing a third chemical site — a surface after a boundary, or a component after a crystal — to save the third landing would be decorating the syntax. It is not done.
There are two chemistry birth sites, lattice site and gap, and one syntactic merge site for the external argument. Interstitial diffusion does not need a vacancy. Substitutional diffusion does. Syntax has no sourced analogue of “the step happens only if a hole is next door.” Inventing one would be a joint mechanism. It is not invented.
If the vacancy mechanism is thought of as the empty site moving, the thing that travels may be the hole, not the participant. That inverts who the participant is. The inversion is a collapse pressure. It is not used. A vacancy is still a defect in the matter. It is not a theta role.
Direct swap and ring exchange are the chemistry pictures DoITPoMS says do not happen in practice. A story in which two arguments trade places is the story those pages set aside.
Grain-boundary travel matters, on the DoITPoMS page, when the boundary path is much faster than the interior, at low temperature, because the boundary is thin. Equilibrium segregation lessens as temperature rises. Non-equilibrium segregation grows as temperature rises and would anneal away. Those are three different temperature stories. None of them is mode, tense, or aspect. Mode is not temperature. Bridge A, which is about how many and whether saturated, is a different strand and is not revised here.
The free surface is the privileged site closest to an exit. At high temperature, evaporation from the surface takes the description outside McLean. Mapping the edge of the clause onto the free surface is how the analogy turns into extraction. The opened Navajo questions do not extract arg1. The English subject move does not extract it from the clause either.
A second phase or a precipitateSecond-phase particle that forms (often on aging) and blocks dislocations. is the participant leaving the host solution. Wikipedia separates that from the segregation discussed here. It breaks the analogy the same way a theta bearer leaving the clause would.
Macrosegregation stays in the casting and still cannot be undone by diffusion, because the distances are too large. The clause-sized unit and the component come apart at that point. Macrosegregation is not the third slot.
The low copy in the English scope argument is still available to interpret. An atom that has hopped is not, on these pages, still sitting at the birth site. Reconstruction is a reason linguists believe the subject started low. It is also a reason not to identify a copy with a diffused atom. A syntactic copy is not a random walk.
Continuous energetics — a barrier, a free-energy difference between lattice and boundary — are not discrete merge. No number is stated, because none is being claimed. The balloon is not on the pages. Lower energy at a boundary is not local levity.
12. The open gap: a theta role is not a piece of matter
HIS. SPECULATIVE. Kept open on purpose. Navajo is the extreme edge case that shows the inner workings. The units that move, on his view, are theta roles, not lexical items. He says chemistry has no super clear correlate for that. It should be investigated and accounted for. It should not be dropped, and it should not be treated as a refutation of the nesting question.
What the chemistry pages actually track is a piece of matter: an interstitial atom, a substitutional atom, a vacancy (a missing atom, which can be narrated as the thing that moves), or a solute atom, ion, or molecule gathering at a boundary or a surface. The pages do not track a role, a label, or a “who” that is distinct from the piece of matter. They do not have a case in which the role changes site and the atom stays, or the role is addressed at a privileged site while no atom relocates.
Vacancy-versus-atom is the closest verbal temptation, and it is the wrong temptation. A vacancy is still a defect in the matter, not a theta role. Saying otherwise would invent the stand-in this section exists to refuse. An electron is not proposed. A site label is not proposed. A grain boundary is not a theta role. Mode is not a temperature. A subject exponent is not a solute.
The pattern that can be matched on the pages opened is geometric, and only for pieces of matter. Ordinary English is likened, cautiously, to an in-solid hop, because a lexical subject really changes site and stays in the clause. Navajo, Belfast’s associate, and Welsh-as-reported occupy a site inside a larger piece without the subject leaving the verb phrase: nesting, no hop. The pattern that cannot be matched is his: a theta role as the unit that is rearranged, with the lexical item not the traveler. No chemical correlate was found. None is proposed.
The next pattern from the earlier picture — whether the audible subject exponent on mode is a privileged site, arg1 addressed there without a climb — is not a silent answer to this gap. HIS. SPECULATIVE. It is a nesting question about an address inside one verb word. It does not assign the exponent to a grain, a surface, a vacancy, or a gap. “Without a climb” is the syntax the drawn analysis already has. It is not a diffusion step under another name.
A later pass would have to open a real chemistry or materials source that distinguishes a role, a site label, or a bookkeeping identity from the atom that carries it, and would still have to show that the source is about that distinction, not about a new name for the atom. Until then the sentence is: chemistry, as opened here, moves matter; the Navajo edge case, as he reads it, is about the role.
13. What stays separate
| Page | Question it owns | What this page does with it |
|---|---|---|
| Derivational movement | Where arg1 is, on the drawn derivation | Inherits the constraint: merged in Spec,vP, unmoved, exponent inside the verb word. Does not redraw the trees. Does not adopt that page’s chemistry half. Does not give Navajo a hop |
| Lexicalization | What is listed | Points at verb base versus verb theme. Does not merge listing with the three slots |
| Causative split | How the causative split is partial | Points at the scout. Does not use ł as a chemical site |
| Zipf | Ranked frequency | Uses the Williams et al. 2015 cut as a boundary, one level off his word-as-atom cut. Leaves the phrasal-verb question open. Does not choose quarks or anyons |
| typed-participant map | Which participant types a mechanism family can wear | Different question. Not merged |
| Solubility windows | Bridge A: windows and saturation | One strand, different from this one. Unrevised |
14. What this page did not add
No Navajo form, gloss, or tree was invented. No Belfast sentence and no Welsh sentence was invented. Koopman and Sportiche 1991 and Woolford 1991 were not opened. French, Italian, and Icelandic were not surveyed. No concentration, barrier, thickness, enrichment ratio, or temperature value was copied. No phrasal-verb exponent was stated. Quarks were not chosen over anyons. The balloon was not turned into a force, and lower free energy was not called levity. The clown-car cab was not turned into a clause type. Bridge A was not revised. A joint field was not announced.
No diffusion step and no segregation step was drawn for Navajo, for the Belfast associate, or for Welsh as reported. The only chemistry mechanisms on the page are the interstitial hop, the vacancy hop, and segregation that stays in the solid, and the hop among them is likened only to ordinary English, cautiously, with the weaknesses left in view.
The open investigation remains open. Theta roles rather than lexical items are his units. Navajo is his edge case for seeing them. A chemistry correlate for a theta role, distinct from a likeness for a moving lexical subject, is not clear. None is invented here. The vacancy is not the role. The electron is not the role. The site label is not the role.
2026-10-04 PT. Whole correspondence SPECULATIVE. Hop only for ordinary English. Belfast English, Welsh as reported, and Navajo stay in the verb phrase: nesting, no diffusion step, no segregation step. Navajo is the edge case. No chemical stand-in for a theta role. No joint field. Not Bridge A.
Corpus note: vpish-molecule-picture-v0.md. The correspondence is speculative. A diffusion or segregation step is likened only to ordinary English. Navajo, Belfast English, and Welsh as reported stay in the verb phrase. No chemical stand-in for a theta role. Not Bridge A. Not a joint field.