Mam Voice System Fragment #
@cite{elkins-torrence-brown-2026} @cite{scott-2023}
Minimalist infrastructure for the Mam verbal domain, connecting the language-specific clause sizes to the ClauseSpine and VoiceHead abstractions.
Variety note: Oblique extraction (=(y)a') data is from SJO (San Juan Ostuncalco) Mam via @cite{elkins-torrence-brown-2026}. Antipassive voice data draws on SJA (San Juan Atitán) Mam via @cite{scott-2023}. These are distinct varieties; shared infrastructure (clause spines, Voice heads) is parameterized where possible.
Key Claims #
- Mam transitive clauses project Voice⁰, which bears [uOblique].
- Mam "aspectless" complements are VoiceP-sized — they lack aspect morphology but still project Voice.
- Mam infinitival complements are VP-sized — they project only V, without Voice, Appl, or v.
- The directional auxiliary (Dir⁰) is a Mam-specific head that also
bears [uOblique] and can host =(y)a'. Dir is modeled as a light verb
(V1 in the verbal template Voice > V1(Dir) > Appl > V2(root)) but is
NOT added to
Catas it is not universal.
Mam agentive Voice head with [uOblique] probe.
In Mam, Voice⁰ probes for an oblique feature on a passing Ā-moved constituent. When an oblique DP moves through Spec,VoiceP, Agree values [uOblique] as [+oblique], which is then spelled out as =(y)a' at PF.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Mam transitive clause spine: full CP with Voice. Projects V, Appl, v, Voice, T, C. =(y)a' possible.
Instances For
Mam aspectless complement spine: VoiceP-sized. Projects V, Appl, v, Voice. Still has Voice → =(y)a' possible.
Instances For
Mam infinitival complement spine: VP-sized. Projects only V — no Voice, no Appl, no v. =(y)a' impossible because there is no Voice⁰ to bear [uOblique].
Instances For
Mam directional auxiliary head (Dir⁰).
Dir is NOT a universal category — it is specific to Mayan languages.
We model it as a language-specific type rather than adding to Cat
(which would cascade to fValue, catFamily, epSemanticType, etc.).
In Elkins et al.'s analysis, Dir⁰ occupies V1 position in the verbal template (Voice > V1(Dir) > Appl > V2(root)). Like Voice⁰, Dir⁰ bears [uOblique] and can host =(y)a'. Both Voice and Dir independently Agree with an oblique passing through their specifier, yielding multiple =(y)a' along the movement path.
- cislocative : Bool
Whether the directional is cislocative (toward speaker) or translocative (away from speaker)
- hasUOblique : Bool
Whether this Dir head carries [uOblique], enabling Agree with passing obliques. Like Voice⁰, Dir⁰ can independently trigger =(y)a' spellout when oblique extraction passes through its domain.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
Dir⁰'s probe features when it carries [uOblique].
Equations
Instances For
Cislocative directional with [uOblique]: movement toward deictic center. Carries the same oblique probe as Voice⁰ (Elkins et al. §3.1, ex. 8).
Instances For
Translocative directional with [uOblique]: movement away from deictic center.
Instances For
Cislocative Dir carries [uOblique].
Translocative Dir carries [uOblique].
Mam transitive spine projects Voice.
Mam aspectless spine projects Voice.
Mam infinitival spine does NOT project Voice.
Mam Voice head carries [uOblique].
Mam Voice assigns a θ-role (agentive).
Vocabulary entry for =(y)a': maps [+oblique] on Voice⁰ to the exponent "=(y)a'". This is the Vocabulary Insertion rule in DM terms.
Equations
- Fragments.Mayan.Mam.eqYaVocab = { features := [Minimalism.GramFeature.valued (Minimalism.FeatureVal.oblique true)], exponent := "=(y)a'", context := some Minimalism.Cat.Voice }
Instances For
The Mam Voice vocabulary: just the =(y)a' entry.
Instances For
Mam passive Voice head: carries [uOblique] just like agentive Voice. The oblique probe is independent of Voice flavor — changing flavor from agentive to non-thematic does not remove [uOblique]. This is why =(y)a' co-occurs with passive -njtz (Elkins et al. §7.2).
Equations
- One or more equations did not get rendered due to their size.
Instances For
Passive and agentive Voice differ in flavor but share the same oblique probe features. =(y)a' (conditioned by features) and -njtz (conditioned by flavor) are structurally independent.
Mam antipassive Voice head.
In antipassive constructions, the agent is still introduced by Voice but receives absolutive case (Set B marking) instead of ergative (Set A). The object is demoted to an oblique introduced by a relational noun. The verb bears the antipassive suffix -(a)n.
Antipassive Voice is NOT a phase head — the subject gets ABS, not ERG. This is the key structural difference from agentive Voice: the clause is syntactically intransitive despite having an agent.
In agent focus / ergative extraction contexts, the antipassive co-occurs with the suffix -ta (@cite{scott-2023}, §2.7.1.3).
Equations
- Fragments.Mayan.Mam.mamAntipassiveVoice = { flavor := Minimalism.VoiceFlavor.antipassive, hasD := true, phaseHead := false }
Instances For
Antipassive Voice assigns a θ-role (the agent is present).
Antipassive Voice is NOT a phase head.
Antipassive and agentive Voice differ in phase-head status but both assign θ-roles. This captures the transitivity alternation: same agent, different case on the subject.
Mam voice system: two-way asymmetrical (agentive/passive).
Unlike Toba Batak's symmetrical pivot system, Mam's agentive voice is the basic form (phase head, overt agent) and passive is derived (non-phase, implicit agent). Voice does not determine pivot for extraction — instead, Voice carries [uOblique] which conditions extraction morphology (=(y)a').
Equations
- One or more equations did not get rendered due to their size.
Instances For
Mam is not a simple active/passive system — it also has antipassive.