Vocabulary Insertion and PF Locality

 

The Domain of Vocabulary Insertion: Constraining PF Interface Locality in Indo-Aryan, Germanic, and Semitic Morphology

Abstract

This paper develops a parametric theory of cross-linguistic variation at the syntax–morphology interface grounded in a restriction on the Domain of Vocabulary Insertion (DVI) at Phonological Form (PF). We argue that the standard typological triad, agglutinative, fusional, and non-concatenative morphology, does not reflect distinct generative mechanisms. Instead, it reduces to three geometric settings of a single interface parameter regulating the maximal syntactic configuration accessible to a Vocabulary Item (VI) within a phase (Φ).

We formalize three regimes: (i) Atomic DVI ($DVI = 1), where insertion applies strictly terminal-by-terminal (Urdu, Saraiki); (ii) Span DVI ($DVI = X^n), where insertion targets contiguous head-adjacent sequences along the functional spine (Germanic strong verb morphology); and (iii) Sub-head DVI ($DVI = 1/n), where a single syntactic terminal is partitioned at PF into parallel autosegmental tiers (Semitic root-and-pattern morphology). The proposal eliminates unconstrained post-syntactic mechanisms such as Fusion and Readjustment Rules, deriving morphological typology from a single phase-bounded locality constraint on PF access to syntactic structure.

1. Introduction: PF Locality and the Exponence Problem

A central unresolved issue in late-insertion approaches to morphology,  most prominently Distributed Morphology (Halle & Marantz 1993, 1994; Harley & Noyer 1999; Embick 2010), concerns the systematic cross-linguistic variation in how syntactic structures are realized phonologically. While narrow syntax is assumed to be universally invariant, governed by Merge and feature-checking operations, the PF interface exhibits striking diversity in exponence patterns.

Languages differ in whether:

each syntactic terminal is realized independently (agglutination),

multiple heads are fused into a single exponent (fusional morphology),

or a single root is distributed across non-linear phonological tiers (templatic morphology).

Standard DM accounts derive these differences through post-syntactic operations such as Fusion or Readjustment Rules. However, such operations are descriptively powerful but theoretically unconstrained: they permit arbitrary restructuring of morphophonological outputs without principled limits on locality or structural adjacency.

This paper argues that these mechanisms are unnecessary. All observed variation follows from a single parameter governing PF locality over syntactic structure:

Domain of Vocabulary Insertion (DVI):

The maximal structurally contiguous domain within a phase Φ accessible to a single Vocabulary Insertion operation.

We propose that morphological typology is not a property of distinct modules but an emergent effect of how PF traverses phase-bounded syntactic structure under this single locality parameter.

2. Formal Architecture of the DVI System

Let Sigma = {H_1, H_2,, H_n} denote a set of head-adjacent terminals within a single phase Φ. Vocabulary Insertion is defined as a mapping:

$$VI : \Sigma \rightarrow \Pi$$

where Pi is the set of phonological exponents.

The DVI parameter constrains the cardinality and configuration of admissible subsets of $\Sigma$ available to VI.

2.1 Atomic DVI ($DVI = 1$)

Under Atomic DVI, VI operates under strict terminal locality:

$$\forall H_i \in \Sigma,\ \exists!\ \pi_i \in \Pi \ \text{such that}\ H_i \leftrightarrow \pi_i$$

​

No adjacency computation is permitted. Each syntactic head is independently spelled out.

2.2 Span DVI ($DVI = X^n$)

Under Span DVI, VI may access contiguous sequences of heads:

$$\exists \sigma \subseteq \Sigma \ \text{such that}\ |\sigma| > 1 \Rightarrow \sigma \leftrightarrow \pi$$

∃σ⊆Σ such that ∣σ∣>1⇒σ↔π

Vocabulary Items may thus compete over structurally adjacent chains along the functional spine.

2.3 Sub-head DVI ($DVI = 1/n$)

Under Sub-head DVI, a single terminal is decomposed at PF into parallel representational tiers:

$$H \leftrightarrow \langle \pi_{seg}, \pi_{voc} \rangle$$

This is not syntactic decomposition but PF-level feature distribution governed by autosegmental association.

3. Atomic DVI: Indo-Aryan as Strict Terminal Realization

Indo-Aryan languages such as Urdu and Saraiki instantiate Atomic DVI. Morphology is strictly compositional, with no evidence of head spanning or template-based fusion.

3.1 Monoclausal Structure

Consider:

• Urdu: likh-vā diyā
• Saraiki: likh-vā ditta

We assume a monoclausal vP phase structure in which causative and light verb elements are independently merged within a single phase domain.

VoiceP
  |
  vP
 /  \
VP   v
 |    |
√ROOT LightV

3.2 Vocabulary Insertion

Given $DVI = 1$, insertion proceeds terminal-by-terminal:

$$\sqrt{LIKH} \leftrightarrow /likh/$$

$$v[\textsc{Caus}] \leftrightarrow /-vā/$$$$Voice/LightV[\textsc{Pfv}] \leftrightarrow /diyā/,\ /ditta/$$

No operation may span multiple heads. Surface linearity is therefore a direct reflection of syntactic adjacency.

3.3 Portmanteau Agreement

Apparent violations (e.g., -ā realizing aspect + φ-features) are resolved by assuming feature bundling internal to a single terminal:

$$[T_{Pfv}] \rightarrow [T_{Pfv} + Agr_{[Masc,SG]}]$$

Since bundling occurs pre-insertion within a single node, DVI is not violated.

4. Span DVI: Germanic Ablaut Without Readjustment

Germanic strong verbs exhibit non-concatenative alternations traditionally attributed to phonological readjustment. We reinterpret these as span-based lexical insertion.

4.1 Span Configuration

For sing → sang:

$$\langle T_{Pst}, v, \sqrt{SING} \rangle \leftrightarrow /sæŋ/$$

The Vocabulary Item targets a structurally contiguous head sequence.

4.2 Blocking Condition

Span insertion is strictly adjacency-dependent. Intervention disrupts licensing:

T → NegP → vP

In such cases, span collapse forces reversion to Atomic DVI:

• did not sing

This follows without phonological rules or derivational repair.

5. Sub-head DVI: Semitic Root-and-Pattern as Tier Segregation

Semitic morphology emerges from PF-level feature stratification rather than templatic storage.

5.1 Structural Input

TP
 |
VoiceP
 |
 vP
 |
√ROOT (KTV)

5.2 PF Tier Split

Under Sub-head DVI:

• Consonantal Tier: /k-t-v/
• Vocalic Tier: /i-e/

$$\sqrt{KTV} \rightarrow /k-t-v/$$

$$[v, Voice] \rightarrow /i-e/$$

5.3 Emergence of Templates

Interdigitation is derived via standard autosegmental association constraints. Templates are not stored objects but emergent alignment configurations at PF.

6. Theoretical Consequences

The DVI system eliminates three core theoretical constructs:

6.1 Against Readjustment Rules

All allomorphy is structurally localized; no distance-based phonological rewriting is permitted.

6.2 Against Fusion

Fusion is eliminated as redundant; span access derives the same empirical effects without tree modification.

6.3 Against Nanosyntax

Phrasal spell-out is strictly disallowed; only head-adjacent domains within phases are accessible.

7. Typological Reduction

Morphological typology collapses into a single parameter space:

System	DVI Setting	Output Type
Urdu/Saraiki	1	Linear agglutination
English	Xⁿ	Span-based alternation
Semitic	1/n	Tier interdigitation

8. Conclusion: Morphology as Interface Geometry

This paper argues that morphological variation does not reflect distinct generative systems but different resolutions of a single PF interface parameter governing locality of Vocabulary Insertion.

The Domain of Vocabulary Insertion reduces morphological typology to a constrained geometry of access over phase-bound syntactic structure. Under this view, morphology has no independent generative capacity; it is the emergent effect of how syntactic objects are licensed for phonological interpretation.

The strongest implication is theoretical simplification: once DVI is adopted, the architecture of Distributed Morphology can be reduced to a strictly local, phase-bounded interface system with no need for post-syntactic derivational operations.

References

Ackema, P., & Neeleman, A. (2004). Beyond morphology: Interface conditions on word formation. Oxford University Press.

Arad, M. (2005). Roots and patterns: Hebrew morpho-syntax. Springer.

Bobaljik, J. D. (2012). Universals in comparative morphology: Suppletion, superlatives, and the structure of words. MIT Press.

Borer, H. (2013). Structuring sense: Volume III: Taking form. Oxford University Press.

Caha, P. (2009). The nanosyntax of case (Doctoral dissertation, University of Tromsø). Munin Research Archive.

Embick, D. (2010). Localism in morphology: Regularity, irregularity, and the structure of words. MIT Press.

Embick, D., & Marantz, A. (2008). Architecture and blocking. Linguistic Inquiry, 39(1), 1–53. https://doi.org/10.1162/ling.2008.39.1.1

Embick, D., & Noyer, R. (2007). Distributed morphology and the syntax/morphology interface. In G. Ramchand & C. Reiss (Eds.), The Oxford handbook of linguistic interfaces (pp. 289–324). Oxford University Press.

Halle, M. (1997). Distributed morphology: Impoverishment and fission. In MIT Working Papers in Linguistics (Vol. 30, pp. 425–449). MITWPL.

Halle, M., & Marantz, A. (1993). Distributed morphology and the pieces of inflection. In K. Hale & S. J. Keyser (Eds.), The view from Building 20: Essays in linguistics in honor of Sylvain Bromberger (pp. 111–176). MIT Press.

Halle, M., & Marantz, A. (1994). Some key features of Distributed Morphology. MIT Working Papers in Linguistics, 21, 275–288.

Harley, H. (2014). On the identity of roots. Theoretical Linguistics, 40(3–4), 225–276. https://doi.org/10.1515/tl-2014-0010

Harley, H., & Noyer, R. (1999). Distributed morphology. GLOT International, 4(4), 3–9.

Harley, H., & Siddiqi, D. (2016). Spanning, fusion, and the architecture of late insertion. Linguistic Inquiry, 47(4), 735–763. https://doi.org/10.1162/LING_a_00228

Kalin, L. (2022). Infixation really is (underlyingly) prefixation/suffixation: Evidence from Amanas. Language, 98(4), 637–681. 

Kastner, I. (2020). Voice at the interfaces: The syntax and morphology of Hebrew verbs. Language Science Press. https://langsci-press.org/catalog/book/271

Marantz, A. (1997). No escape from syntax: Don’t try morphological analysis in the privacy of your own lexicon. In A. Dimitriadis, L. Siegel, C. Surek-Clark, & A. Williams (Eds.), Proceedings of the 21st Penn Linguistics Colloquium (pp. 201–225). University of Pennsylvania Working Papers in Linguistics.

Marantz, A. (2001). Words. Manuscript, Department of Linguistics and Philosophy, Massachusetts Institute of Technology.

McCarthy, J. J. (1981). A prosodic theory of nonconcatenative morphology. Linguistic Inquiry, 12(3), 373–418.

Merchant, J. (2015). How much context is enough? Two cases of span-conditioned stem allomorphy. Linguistic Inquiry, 46(2), 273–303. 

Nevins, A. (2011). Multiple agree with clitics. Natural Language & Linguistic Theory, 29, 939–971.

Pylkkänen, L. (2008). Introducing arguments. MIT Press.

Ramchand, G. C. (2008). Verb meaning and the architecture of the clause. Cambridge University Press.

Siddiqi, D. (2009). Syntax within the word: Economy, allomorphy, and argument selection in Distributed Morphology. John Benjamins.

Starke, M. (2009). Nanosyntax: A short primer. Nordlyd, 36(1), 1–6. 

Svenonius, P. (2016). Spans and words. In H. Harley & D. Siddiqi (Eds.), Morphological metatheory (pp. 201–222). John Benjamins.

Wurmbrand, S. (2001). Infinitives: Restructuring and clause structure. Mouton de Gruyter.