Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
When PF instructs syntax
Transcript of When PF instructs syntax
On syntax-PF mapping issues
update chunk definition in phonology
reduction to just one chunk-defining device
Chunk definition in phonology
How are phonologically relevant chunks (i.e. domains of phonological computation) of the linear string defined?
since SPE: two competitors
Representationally: #s in SPE,
units of the Prosodic Hierarchy
Procedurally: cycles, today phases
Since Lexical Phonology, post-lexical phonology is supposed to be non-cyclic.
At and above the word level, chunk definition is ONLY representational, i.e. done by the Prosodic Hierarchy.
Phase theory obliterates this idea: it defines phonologically relevant chunks above the word level
1. It is redundant
2. It is made of diacritics
Adapting phase theory to the demands of phonology
As it stands, phase theory is unable to describe all phonologically relevant chunks, which are often smaller and more fine grained than what phase theory can delineate.
Phase theory in a nutshell
- Phases are domains of SYNTACTIC computation
-Phase heads (usually, v and C, maybe D) are the loci of first-Merge of uninterpretable phi-features
- PHASE IMPENETRABILITY CONDITION (PIC)
In a phase αP with head H, the domain of H is not accessible to operations outside P ; only H and its edge are accessible to such operations.
(Chomsky 2000: 108)
- Phases and PICs are isomorphic:
every phase triggers a PIC, and every PIC is linked to a phase.
1. The phase skeleton is defined in syntax (phasehood: which nodes are phase heads, i.e. trigger spell-out to PF?)
2. For a given phase, the PIC is module-specific
Within a given language, it is decided for every phase whether or not
- it is associated to a PIC in syntax
- it is associated to a PIC at PF
- it is associated to a PIC at LF
Cheng & Downing: what does phonology know about syntax?
what matters are not DOMAINS but EDGES
Penultimate vowel lengthening: marks the limits of the prosodic phrase.
Tone spreading within a domain
Abruzzese has person-driven auxiliary selection
so magnate /I am eaten
si magnate /You are eaten
a magnate /(s)he has eaten
'I/you have eaten, he has eaten'
so mmagnate /I am eaten
si mmagnate/You are eaten
jè mmagnate /He is eaten
'I am/you are/(s)he is eaten
The edge-based approach to prosodic phrasing
Case study I
Simple sentences in Zulu & Chichewa
Case study II
Restrictive relative clauses
in Zulu & Chichewa
Conclusion: phases cannot be
mapped directly into prosodic domains
The head of the RC and the RC belong to the same prosodic domain (contra what phase theory predicts]
The V belongs to the same prosodic domain as IO and DO
a. ALIGNR[PHASE, INTPH] (ALIGNR-PHASE): Align the right edge of every phase (P/CP) with the right edge of an Intonation Phrase (IntPh).
b. ALIGNR[INTPH, PHASE] (ALIGNR-INTPH): Align the right edge of every Intonation Phrase (IntPh) with the right edge of a phase (P/CP).
The Edge-based approach to prosodic phrasing
Cheng & Downing's solution
We still need the syntactic chunk to be translated into phonological representations in the form of a unit of the Prosodic Hierarchy
Passives are obtained by means of Raddoppiamento Fonosintattico (RF)
what happens at PF?
transitive v has a PIC at PF
Reaction/adaptation of the established Prosodic Hierarchy to phase theory:
Prosodic islands make prosodic constituency isomorphic with phases: FIRST a phase defines the chunk, THEN this chunk is translated into phonological representations in the form of a unit of the Prosodic Hierarchy.
==> abandon of THE fundamental claim of Prosodic Phonology: non-isomorphism.
Dobashi (2003), Piggott & Newell (2006), Kratzer & Selkirk (2007), Ishihara (2007) and Kahnemuyipour (2009); Elordieta (2008:274ff) offers an informed survey.
Theory can't afford to do the same labor twice
Phases are independently needed in syntax
but the prosodic hierarchy is not independently needed in phonology
Pak (2008:42ff), Samuels (2009: 284ff), Seidl (2001)
1. It is redundant
Just like #, it is made of diacritics (you know everything about this)
In a modular environment, diacritics do not qualify since computational systems (modules) can only understand, parse and process their own proprietary vocabulary.
[Scheer (2008a, 2011:§402, 2012a:§93, 2012b)]
How it works
Cheng & Downing (2007, 2009, 2012)
Leiden University Centre for Linguistics
D'Alessandro & Scheer (2013). Phase head marking. In Hisao Tokizaki and Yoshi Dobashi (eds), Linguistic Analysis. Special issue on Universal Syntax and Parametric Phonology 38:4, 305-330.
D'Alessandro & Scheer. to appear. Modular PIC. Linguistic Inquiry
Unaccusatives in Abruzzese are
*syntactically* like passives, but phonologically like actives
Domain specificity in Cognitive Science
Segal (1996:145), Gerrans (2002:261), Cosmides & Tooby (1992), Hirschfeld & Gelman (eds.) (1994), Fodor (2000:58ff)
How it works
No isomorphism between spellout domains and phonological domains, but we still need phases
Problem: Unaccusatives do not have RF
Virtual conceptual necessity
Simplicity, elegance, economy
Postulate only what is necessary
Strong Minimalist Thesis
Legibility: PF and LF must be able to "read" what NS produces
The Minimalist Program
“Language is an optimal solution
to legibility conditions”
PF (and LF) must instruct syntax
we need to take phonology more seriously
joint work with
Prosodic hierarchy must go (Tobias's orders) because:
are phases really independently needed?
Separate the PIC from Spell-Out
Y/T models: syntax is at the core.
The output of syntax goes to PF/LF
MAPPING of structural units to phonological/prosodic units
phase heads: locus of Merge of uninterpretable features
Do we ever "see" phase heads at PF??
Complementizer doubling in Southern Italian dialects
just delimiting Focus?
in a split-CP framework
is the Force complementizer - no need for it in
marks the phase head