Linguistics primer map: Valency, theta roles, and thematic relations
What’s claimed and how well supported
- Labels on this page
- SPECULATIVE ×1
- Strongest label used
- SPECULATIVE
- Main caveat (quoted)
- ◆ SPECULATIVE (do not treat as established linguistics): any mapping from this map onto metal/semiconductor doping.
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.
Project: Metal Doping × Theta Roles
Scope: Linguistics side only.
Date: 2026-09-06
Explicit scope note: This file contains no materials-science or doping analogies. It maps the linguistic concepts needed for later (speculative) bridging. Four-place / quadrivalent predicates are rare and often disputed as true core-argument structures; Grant’s primary takeaway states that any later analogy work that extends three-argument patterns must treat four-place structures as optional elaborations, not as robust universal building blocks.
Primary project source (Grant): ../theta-roles-takeaway.txt — Four-Place Verbs: Syntax and Semantics (Aug 2026 notes). Cast inventory: grant-linguistics-cast.md (not published here).
- 1. Lead concept: Valency and argument saturation
- 2. Four-place verbs in detail
- 3. Thematic relations (descriptive inventory)
- 4. Theta roles and the θ-criterion (generative formal saturation)
- 5. Dowty (1991): Proto-roles (critique of discrete inventories)
- 6. Open questions / common confusions
- 7. Source list
- 8. One-page takeaway map (linguistics only)
1. Lead concept: Valency and argument saturation
1.1 What valency is
In the Tesnière tradition (dependency grammar / valencyArity: how many actants a predicate takes; formal vs semantic valency can diverge. theory), a verb is like a “hooked atom”: its valency is the number of actants (arguments / complements) it can or must govern. The subject counts as an actant on a par with objects—unlike early phrase-structure subcategorization frames that sometimes omitted subjects.
- Semantic valency: how many participant roles the event/state conceptually involves.
- Formal (syntactic) valency: how many of those participants are realized as core arguments (obligatory or strongly selected syntactic dependents), as opposed to adjuncts / obliques.
A verb may have four semantic participants but only two or three syntactic arguments. That mismatch is central to honest argument-structure work (Grant §5; see §6 below).
1.2 Valency hierarchy (0–4)
Standard descriptive hierarchy (Grant §2; Wikipedia “Valency (linguistics)”; Tesnière tradition):
| Valency | Name | Typical arguments | English example |
|---|---|---|---|
| 0 | Avalent / impersonal | none (or dummy subject) | It rains |
| 1 | Monovalent / intransitive | subject | He sleeps |
| 2 | Divalent / transitive | subject + object | She kicked the ball |
| 3 | Trivalent / ditransitive | subject + IO + DO | He gave her a book |
| 4 | Quadrivalent / tritransitive | four arguments | I bet him five dollars on the horse |
Notes:
- Four-place verbs (also called quadrivalent, tetravalent, tritransitive, quadradic) sit at the extreme end of the hierarchy commonly discussed for natural language.
- Claims of five-or-more places are extremely rare and are usually reanalyzed as adjuncts or complex predicates rather than true core arguments (Grant §2).
- Cross-linguistically, ~3 core arguments is the practical maximum for most languages; true underived four-place verbs are scarce (Grant §5–6).
1.3 Formal vs semantic valency (critical distinction)
| Semantic valency | Formal / syntactic valency | |
|---|---|---|
| Counts | Participants entailed by the event meaning | Dependents selected / governed as core arguments |
| Example | Commercial exchange: buyer, seller, goods, price | English sell often patterns as ditransitive + optional for-PP |
| Consequence | Four “roles” in the scene | Possibly only 2–3 slots that must be filled for well-formedness |
Grant’s formulation: A verb may have four semantic participants but only three (or fewer) syntactic arguments; the extra participant is realized as an adjunct or incorporated into a complex predicate.
This is the same territory where generative θ-theory talks about saturation—but valency theory states it in dependency / actant terms first.
2. Four-place verbs in detail
2.1 English candidates and the adjunct vs argument debate
Most cited English candidate: bet (Grant §3.1):
- I₁ bet him₂ five dollars₃ on “The Daily Arabian”₄
- I₁ bet you₂ two dollars₃ that it will rain₄
Other financial-exchange verbs sometimes treated similarly:
- Pat₁ sold Chris₂ a lawnmower₃ for $20₄
- Chris₁ paid Pat₂ $20₃ for a lawnmower₄
Syntactic debate (flag: unsettled):
- Many analyses treat the fourth element (that-clause, on-PP, for-PP) as an adjunct, because it can often be omitted without ungrammaticality (I bet you two dollars).
- True core four-place status in English is therefore controversial.
- The fourth participant is frequently a PP or clausal complement, not a bare NP.
- Parallel point in the literature: Jackendoff-style semantic analyses may treat commercial events as four-participant at the conceptual level while grammar demotes price/source to adjunct (semantic ≠ syntactic arity).
Working stance for this project: treat English bet/sell/pay as textbook candidates, not as settled proof of a robust universal four-place class.
2.2 Languages with clearer morphological marking
Where polypersonal agreement or rich case morphology indexes multiple arguments on the verb, four-argument encoding can be clearer (Grant §3.2):
| Language / family | What is reported | Status |
|---|---|---|
| Abaza (Northwest Caucasian) | Causative of a ditransitive can yield a verb with four pronominal prefixes (causer + causee + theme + recipient); Abaza verbs can cross-reference up to four arguments | Strong morphological evidence for derived 4-place |
| Ukrainian, Russian, other Slavic | Some semantic–syntactic studies describe four-valent verbs (motion/transfer; instrument + source/goal + object) | Descriptive; fourth often peripheral |
| Caucasian / Georgian (reported) | Distinction of four grammatical relations (Subject + Primary Object + two Secondary Objects) in some descriptions | Needs language-specific verification |
Abaza detail (supporting literature): polysynthetic head-marking; absolutive vs oblique (ergative / IO) index series; valency-increasing morphology (causative, applicatives) adds indexed participants (Arkadiev & colleagues; UD Abaza treebank notes; O’Herin on applicatives).
2.3 Derived four-place structures
True underived four-place verbs are scarce. Most attested cases arise via valency-increasing operations (Grant §3.3):
- Causativization of a ditransitive (adds a causer).
- Applicativization (adds beneficiary, instrument, location, etc., as a core argument).
- Double causatives (rare and often restricted).
In practice, languages often avoid pure four-core-argument verbs by demoting one participant to an oblique or adjunct.
2.4 Typical semantic role combinations in four-place frames
Not a fixed universal inventory—recurring combinations (Grant §4):
- Agent / Causer
- Recipient / Beneficiary / Goal
- Theme / Patient / object transferred
- Source / Instrument / Price / Location / Proposition
Common semantic classes:
| Class | Typical frame |
|---|---|
| Betting / wagering | Agent + Opponent + Stake + Outcome/Event |
| Commercial exchange | Agent + Recipient + Goods + Price (buy, sell, trade, purchase) |
| Translation / transfer with two endpoints | Agent + Theme + Source + Target (translate X from Y to Z) |
| Motion / displacement with instrument | Agent + Theme + Instrument + Path/Goal (esp. some Slavic descriptions) |
The fourth argument is almost always more peripheral in the thematic hierarchy than the classic three (Agent–Theme–Recipient), which is why it often surfaces as an oblique.
2.5 Project-relevant caution (from Grant’s own summary)
Four-place verbs exist but are rare, often derived, and frequently disputed as true core-argument structures in natural languages… This scarcity explains why four-argument verbs constitute a relative “dead end” for stable, community-tested structures in natural language, while remaining straightforward in formal or constructed systems. Any attempt to extend three-argument patterns (for example in materials or alloy analogies) must therefore treat four-place structures as optional elaborations rather than as robust, universally attested building blocks.
(That materials/alloy mention is Grant’s own caution about later analogy work; this linguistics file still does not develop doping analogies.)
3. Thematic relations (descriptive inventory)
Thematic relations (also: semantic roles, participant roles, deep cases in Fillmore’s early sense) describe how a participant relates to the event/state denoted by the predicate.
They are a descriptive semantic inventory, not by themselves a biunique syntactic filter.
3.1 Common labels (illustrative, not exhaustive)
| Relation | Rough characterization | Example |
|---|---|---|
| Agent | Volitional initiator / doer | Kim opened the door |
| Patient | Entity undergoing affected change | Kim broke the vase |
| Theme | Entity that moves, is located, or is transferred (boundaries vs Patient disputed) | Kim sent the letter; The book is on the table |
| Experiencer | Holder of a mental/perceptual state | Kim fears spiders |
| Stimulus / Cause (of psych events) | What triggers the experience | Spiders frighten Kim |
| Goal / Recipient | Endpoint / receiver | Kim gave the book to Sam / gave Sam the book |
| Source | Starting point | Kim took the book from Sam |
| Instrument | Means used | Kim opened the door with a key |
| Locative | Location | Kim slept in the tent |
| Benefactive | For whose benefit | Kim baked a cake for Sam |
| Price / Extent / Proposition (esp. exchange & bet) | Often fourth, often peripheral | for $20; that it will rain |
Historical anchors: Gruber (1965) thematic relations; Fillmore (1968) Case Grammar “deep cases”; Jackendoff (1972, 1983, 1990) conceptual-structure roles.
3.2 Theme vs Patient (classic confusion)
There is no field-wide consensus. Typical textbook practice:
- Patient: affected undergoer of change (especially change of state).
- Theme: entity in motion, location, or transfer (sometimes extended to stative “located” entities).
Some frameworks collapse them; others keep them distinct. Dowty (1991) notes that whether Theme equals Patient remains unsettled—one motivation for proto-roles instead of a fixed discrete inventory (§5).
4. Theta roles and the θ-criterion (generative formal saturation)
4.1 Theta roles as syntactic argument slots
In Government–Binding / Principles & Parameters generative grammar, a theta role (θ-role) is a formal argument position associated with a predicate’s lexical entry (often listed in a θ-grid).
Textbook distinction (Carnie; Haegeman instructor notes):
| Thematic relations | Theta roles | |
|---|---|---|
| Domain | Semantic participation in the event | Syntactic slots that must be saturated |
| How many per NP | An NP may bear multiple thematic relations | An argument bears exactly one θ-role (under classical θ-criterion) |
| Who has them | Almost any NP can express thematic content | Only arguments (not free adjuncts) receive θ-roles from the verb |
| Naming | Labels like Agent, Goal… | Often named by the most prominent thematic relation in that slot |
Classic illustration: Susan gave Bill the paper — Susan may be both Agent and Source thematically, but receives one external θ-role (the “agent” slot).
4.2 The θ-criterion (Chomsky 1981)
θ-criterion: Each argument bears one and only one θ-role, and each θ-role is assigned to one and only one argument.
— Chomsky, Lectures on Government and Binding (1981), p. 36
Functions in classical GB:
- Enforce biunique saturation of the verb’s argument structure.
- Distinguish real arguments from dummies (it, there) and from adjuncts.
- Help track argument identity under movement (with the Projection Principle).
Related: Extended Projection Principle (EPP)—clauses need subjects even when the subject is not a θ-marked argument of the main verb (expletives; raising).
4.3 θ-grids (Haegeman / Carnie style)
Example sketch for give:
| Agent (Source) DP | Theme DP | Goal PP/DP |
|---|---|---|
| i | j | k |
Applied to [Susan]ᵢ gave [the food]ⱼ [to Biff]ₖ.
Haegeman (1991/1994) and Carnie (Syntax: A Generative Introduction, ch. on Theta Theory) use box notation for θ-grids; Carnie explicitly separates thematic relations (semantic) from θ-roles (syntactic slots).
4.4 GB → Minimalist reframing (flag: theory-internal change)
In GB, θ-theory + θ-criterion are core filters at D-structure / LF.
In the Minimalist Program, several GB cornerstones (including aspects of the Projection Principle / θ-criterion package) are revised or eliminated in successive formulations (Chomsky 1995 and later chapters). Some later proposals treat θ-roles as features to be checked, or allow multiple θ-assignment under movement (e.g. Hornstein-style control-as-movement)—controversial and not needed for a primer map. Record only: classical biunique θ-criterion is GB-era; MP does not preserve it unchanged.
4.5 How valency and θ-theory relate (without collapsing them)
| Valency (Tesnière lineage) | θ-theory (Chomsky GB lineage) |
|---|---|
| Counts actants a verb governs | Requires each θ-slot filled by exactly one argument |
| Strong on argument vs adjunct diagnostics | Strong on biunique mapping & movement |
| Hierarchy 0–4 is descriptive typology | θ-grids usually list 1–3 roles for English verbs |
| Four-place = extreme / rare | Four θ-roles in one underived grid = rare / disputed |
Bridge phrase for this project: both traditions care about saturation—whether the predicate’s required participants are present—but valency hierarchy (esp. rarity of 4) is the lead frame from Grant’s takeaway; θ-roles/θ-criterion supply the generative formalization of biunique assignment.
5. Dowty (1991): Proto-roles (critique of discrete inventories)
Primary paper: David Dowty, “Thematic Proto-Roles and Argument Selection,” Language 67(3): 547–619 (1991).
Open PDFs: Rochester copy; Berkeley copy.
Secondary guide: Beth Levin (2019), “On Dowty’s ‘Thematic Proto-roles and Argument Selection’” — Stanford PDF.
5.1 Problem Dowty attacks
Traditional discrete roles (Agent, Patient, Source, …) are hard to define with necessary-and-sufficient conditions; inventories fragment; argument-indexing via a short universal role list is empirically fragile—yet GB’s θ-criterion seemed to need something like biunique role types.
5.2 Proposal
Only two cluster concepts:
Proto-Agent entailments (Dowty 1991:572): volition; sentience/perception; causing change in another participant; movement relative to another participant; (independent existence).
Proto-Patient entailments: undergoes change of state; incremental theme; causally affected; stationary relative to another’s movement; (existence not independent of the event).
Argument Selection Principle: the argument with the most Proto-Agent entailments → subject; most Proto-Patient entailments → object. Near-ties → alternations (e.g. fear / frighten).
Traditional labels (Instrument, Experiencer, …) emerge as recurring subsets of entailments—not atomic primitives.
5.3 Relevance to four-place / exchange verbs
Dowty discusses commercial buy/sell, spray/load, and psych verbs as argument-selection problems. Price, partner, and dual Themes in exchange events are exactly where semantic multi-participant scenes outrun neat discrete role lists—and where English often demotes a participant to oblique (formal valency < semantic valency).
6. Open questions / common confusions
- Theme vs Patient — Distinct or collapsed? No consensus (§3.2).
- Argument vs adjunct for English bet / sell / pay / rent fourth phrases — Core four-place or optional elaboration? Controversial; lean adjunct unless morphologically forced elsewhere.
- Thematic relation ≠ θ-role — Same labels, different jobs (semantic content vs syntactic slot). Generative textbooks often warn against equating them; many papers still use the terms interchangeably.
- Individual vs type θ-roles — Marantz / van Riemsdijk & Williams style “hitter role” vs traditional Agent: individual roles trivially satisfy biuniqueness but lose cross-verb generalizations (Dowty’s critique).
- Jackendoff vs GB — Jackendoff allows multiple thematic relations per argument and roles “in the verb”; classical θ-criterion forbids dual θ-assignment to one argument. Different research programs.
- Formal vs semantic valency — Four participants in the scene ≠ four core syntactic arguments. Must keep separate for any later analogy work.
- Derived vs underived 4-place — Morphological causatives/applicatives can create four indexed arguments (Abaza); English candidates are mostly not of that type.
- Is there a universal four-place class? — Grant: no robust universal class; evidence weak in most languages. Speculative claims of widespread tetravalency should be flagged.
- MP status of θ-criterion — Classical formulation is GB; do not assume it is unchanged in current Minimalism.
- Bierwisch intrinsic vs extrinsic thematic roles — Intrinsic (roles emerge from decomposed SF) vs extrinsic (fixed universal role list + Neo-Davidsonian conjuncts). Meta-theoretical; useful for advanced reading (Bierwisch 2006 PDF).
7. Source list
Primary project source
- Grant Golland (comp.), Four-Place Verbs: Syntax and Semantics — Valency, Argument Structure & Cross-Linguistic Notes (Aug 2026). Local:
../theta-roles-takeaway.txt. Cast: grant-linguistics-cast.md (not published here).
Primary / classic linguistics
- Tesnière, Lucien. 1959/1966. Éléments de syntaxe structurale. (Foundational valency / actant theory.)
- Chomsky, Noam. 1981. Lectures on Government and Binding. Foris / Mouton. θ-criterion at p. 36.
- Chomsky, Noam. 1995. The Minimalist Program. MIT Press. (Revisions to GB θ-package—use cautiously.)
- Dowty, David. 1989. “On the Semantic Content of the Notion ‘Thematic Role’.” In Chierchia, Partee & Turner (eds.), Properties, Types and Meaning II. Kluwer.
- Dowty, David. 1991. “Thematic Proto-Roles and Argument Selection.” Language 67(3): 547–619.
- https://www.cs.rochester.edu/u/james/Papers/Dowty.1991.pdf
- https://linguistics.berkeley.edu/~syntax-circle/syntax-group/dowty91.pdf
- JSTOR: https://doi.org/10.2307/415037
- Fillmore, Charles J. 1968. “The Case for Case.” In Bach & Harms (eds.), Universals in Linguistic Theory. Holt, Rinehart and Winston.
- Gruber, Jeffrey. 1965. Studies in Lexical Relations. MIT dissertation.
- Jackendoff, Ray. 1972. Semantic Interpretation in Generative Grammar. MIT Press.
- Jackendoff, Ray. 1990. Semantic Structures. MIT Press.
- Baker, Mark. 1988. Incorporation. University of Chicago Press. (UTAH.)
Textbooks / handbook chapters (secondary but authoritative)
- Haegeman, Liliane. 1991/1994. Introduction to Government and Binding Theory. Blackwell. (θ-grids, θ-criterion.)
- Carnie, Andrew. Syntax: A Generative Introduction (esp. “Constraining X-bar: Theta Theory” / theta grids chapters; 4th ed. Wiley-Blackwell). Author page: https://www.andrewcarnie.org/textbook4e
- Levin, Beth & Malka Rappaport Hovav. 2005. Argument Realization. Cambridge University Press.
- Van Valin, Robert D. & Randy J. LaPolla. 1997. Syntax: Structure, Meaning and Function. CUP. (RRG macroroles—generalized roles parallel to Dowty.)
Open / accessible secondary
- Levin, Beth. 2019. “On Dowty’s ‘Thematic Proto-roles and Argument Selection’.” https://web.stanford.edu/~bclevin/dowty19fin.pdf
- Bierwisch, Manfred. 2006. “Thematic Roles – Universal, Particular, and Idiosyncratic Aspects.” In Bornkessel et al. (eds.), Semantic Role Universals and Argument Linking. de Gruyter. https://www.leibniz-zas.de/fileadmin/media/Dokumente/Personen/Bierwisch/11_Bierwisch_2006_Thematic_Roles.pdf
- Wikipedia (orientation only, verify against primaries): Valency (linguistics); Theta role; Theta criterion.
- Bárány, András. 2017. EGG summer-school slides on argument structure & theta roles. https://andras.barany.at/egg2017/07-26-arguments-and-theta-roles.pdf
Cross-linguistic four-place / Abaza (supporting)
- Descriptive notes on Abaza polypersonal indexing and causative/applicative valency increase (Arkadiev et al.; UD Abaza treebank overview: https://aclanthology.org/2023.udw-1.1.pdf).
- O’Herin, Brian. 2001. Work on Abaza applicatives (cited in NW Caucasian surveys).
8. One-page takeaway map (linguistics only)
VALENCY (Tesnière) θ-THEORY (GB) THEMATIC RELATIONS
arity 0–4 θ-grid + θ-criterion Agent, Theme, Goal…
actants vs adjuncts biunique saturation descriptive semantics
formal ≠ semantic valency slots, not role atoms may many-to-one with θ-slots
\ | /
\ | /
---- FOUR-PLACE EDGE CASE (rare, often derived, often oblique) ----
English bet/sell/pay: disputed core status
Abaza etc.: morphological 4-place via causative/applicative
Project rule: treat 4-place as OPTIONAL ELABORATION
SPECULATIVE (do not treat as established linguistics): any mapping from this map onto metal/semiconductor doping. That work belongs in a separate bridge document, not here.
Compiled 2026-09-06 for Metal Doping × Theta Roles. Prefer Grant’s four-place/valency takeaway + Dowty/Chomsky/Haegeman–Carnie over blog summaries. Flag controversies rather than resolving them by fiat.
Corpus note: linguistics-theta.md