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
The phonological world of Tobias Scheer
Transcript of The phonological world of Tobias Scheer
marc van oostendorp
'This book presents a development of Jean Lowenstamm's idea that phonological constituent structure can be reduced to a strict sequence of non-branching Onsets and non-branching Nuclei.'
A Lateral Theory of Phonology, 1st sentence
1. developing ideas
2. reduction to simple principles
3. inspiration by syntactic theory
1. Science as a cumulative enterprise
- many of the ideas about cvcv are inspired by the work of Jean Lowenstamm
- especially the guide to morphosyntax-phonology interface theories is an example of cumulative work: extracting the common ground from all of the history of phonology (since 'Trubetzzkoy's Grenzsignale'
being truly cumulative does not just mean cherrypicking other people's ideas for the sake of an individual analysis (everybody does that), but also evaluating them critically, and sticking to them if one believes they are right
and trying to make a coherent framework out of them larger than 1 paper
Reading Jean Lowenstamm
'cv as the only syllable type'
'the beginning of the word'
together these papers propose, building on previous work, that the representational assumption of CV units is necessary in order to give the right generalisations on e.g. long vowels, consonant clusters, and the difference between languages that only allow TR clusters and those that are more liberal.
What is a CV Unit?
The CV unit has some internal structure. Where does it come from? Why not VC? Or (V)(C)(V)?
(the downside of cumulativity)
I do not find a real answer to this question: it is CV because it works, and because that is a remnant of syllable theory
I find this particularly damaging, because it makes it difficult to evaluate the claim 'that there is no syllabic arborescence'
2. Reduction to simple principles; 'minimalism'
an important part of Scheerian methodology is to get rid of 'Pink Panthers': uninterpretable 'diacritic' phonological objects.
also Scheer's ideas about the interface serve to keep things 'pure': morphosyntax and phonology should be completely separate, no 'modularity offenders' are allowed.
An important point for Scheer's modularity view is that two modules which are by necessity completely distinct should not share any vocabulary. Mapping is done by a 'lexicon'. (which is not itself a module)
This is a very radical view, and a consequence is that Scheer considers basically all other interface theories 'modularity offenders'
Scheer's lexicon mapping syntax to phonology
- nodes in the tree are mapped to phonological objects
- edges (of phases) are mapped to CV units
there is no explanation why it is never the other way round (nodes are assigned empty CV slots, edges are assigned phonemic material).
in that sense, CV slots are still Pink Pantheresque objects.
(also because remember that we do not know what the status of 'CV units' are)
3. Inspiration by syntactic theory
CVCV is clearly an offspring of Government Phonology, which incorporated syntactic notions (government, ECP) into phonology
It distances itself from 'classic' GP by abandoning many parallels (because of radical modularity)
but at the same time, it stays methodologically close to (Chomskyan) syntactic theory
combining cvcv with minimalism: the paradox
while cvcv (and gp) is strongly representationally oriented,
minimalism has a very strong emphasis on computation ('computation is king')
the 'lateral theory of phonology' incorporates both (at least at the interface; it's not clear what role computation plays inside the phonology)
An example piece of data
Slavic has words which start with rd-, whereas English does not.
On the other hand, not ALL logical options of CC are allowed (unlike in Semitic)
Status of evidence
Assume that C1 in 'anything goes' languages has a special status ('extraprosodic'):
it is adjoined directly to the Prosodic Word
or attached in a later stage to the onset, violating sonority
or left astray
Scheer criticizes such analyses for being unrestrictive, unmotivated and (at least in the former case) using diacritics
A 1-slide introduction to CVCV phonology for the already initiated, using a very non-Scheerian word as an example
licensed because final
inside TR cluster
licensed by following full V, modulo TR
Special status of V
Special status of TR
A 1-slide introduction to the interface
licensed by TR
disclaimer: I use 'license' in a nontechnical sense
as far as I am able to tell, in the Scheerian world we have the following types of evidence:
static distributions (in corpora), phonological alternations, historical change (including loanword adaptation), geographic distribution
(i am not sure that i know what motivates these choices)
Experimental work (e.g. Shaw et al. 2009) has established that Moroccan speakers time TRV sequences differently from English speakers
They suggest this is evidence that T is syllabified separately (in an empty-headed syllable) in MA, but in a TR cluster in English
It is not clear whether this result is meaningful in CVCV interpretations (maybe it means that interconsonantal gvt is a parameter that is only switched on if no other parses are available)
I guess one could take this as an instigation into further scrutiny of the empty V inside TR clusters
Here are a few topics that future Scheerologists might want to study:
- the identity of the CV unit (internal structure)
- the relation between derivation and representation
- the nature of evidence (experimental and otherwise)