A substance-free theory of laryngeal features

Transcript

1. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism . . A substance-ee theory of laryngeal features Pavel Iosad University of Ulster p.iosad@ulster.ac.uk 22 April 2013 University of Edinburgh / Oilthigh Dhùn Èideann Pavel Iosad A substance-ee theory of laryngeal features
2. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Plan Brief introduction Pavel Iosad A substance-ee theory of laryngeal features
3. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Plan Brief introduction Substance-ee phonology: what, why, and how Pavel Iosad A substance-ee theory of laryngeal features
4. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Plan Brief introduction Substance-ee phonology: what, why, and how Decoupling phonology om substance, or, why laryngeal realism is too realistic Pavel Iosad A substance-ee theory of laryngeal features
5. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Plan Brief introduction Substance-ee phonology: what, why, and how Decoupling phonology om substance, or, why laryngeal realism is too realistic Where’s phonetics in phonology? Pavel Iosad A substance-ee theory of laryngeal features
6. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Brief introduction: about Undergraduate degree om Moscow State University, Russia PhD (defended in February), University of Tromsø Thesis: Representation and variation in substance-ee phonology: a case study in Celtic In-depth analysis and comparison of the phonology of two closely related varieties (Welsh and Breton) Pavel Iosad A substance-ee theory of laryngeal features
7. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Brief introduction: research interests Theoretical phonology Phonological representations: substance-ee phonology Division of labour Phonetics–phonology interface: the interaction of phonological computation and ‘gradient’ implementation Phonology–morphosyntax interface, especially stratal/cyclic models Celtic languages Welsh primary specialism Also done work on Breton, Irish, Scottish Gaelic Also done work on Romance, Slavic, Germanic… Historical linguistics Historical phonology Historical language contact Pavel Iosad A substance-ee theory of laryngeal features
8. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism What is substance-ee phonology? Why substance-ee phonology? The workings of the theory Outline . . . 1 Substance-ee phonology: what, why, and how . . . 2 Celtic laryngeal phonology vs. laryngeal realism . . . 3 Substance-ee laryngeal realism Pavel Iosad A substance-ee theory of laryngeal features
9. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism What is substance-ee phonology? Why substance-ee phonology? The workings of the theory Key postulates Phonology is an independent module of grammar ‘Integrative module’ in Jackendovian terms: a computational mechanism mapping om phonological strings to phonological strings Several areas traditionally of concern to phonologists are taken out of the phonological part of the grammar Mapping between phonology and substance is not phonology sensu stricto Substantive considerations (‘markedness’, perceptibility, ease of production) are not phonology sensu stricto Influence of substantive factors mediated by acquisition / diachrony (Ohala 1981; Blevins 2004) Pavel Iosad A substance-ee theory of laryngeal features
10. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism What is substance-ee phonology? Why substance-ee phonology? The workings of the theory Key references ‘Concordia school’: Hale and Reiss (2000, 2008); Hale, Kissock, and Reiss (2007); Reiss (2003, 2007) Substance-ee computation, phonology – phonetics mapping is a quick and easy transduction Substance-ee as understood here: Morén (2006, 2007); Uffmann (2007, 2013); Blaho (2008); Youssef (2010); Iosad (2012a, 2012b, 2012c); Odden (2013) Substance-ee computation and a non-trivial mapping between phonological representations and substance Pavel Iosad A substance-ee theory of laryngeal features
11. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism What is substance-ee phonology? Why substance-ee phonology? The workings of the theory Motivating substance-ee phonology Modularity is important for generative theorizing, which is predicated on a type of knowledge that is specific to language The locus classicus is Fodor (1983), but see also Jackendoff (2000, 2002) Contrast parallel architectures in the mould of Rumelhart and McClelland (1986) And note their historical connection to OT (Smolensky and Legendre 2006; Scheer 2011) Pavel Iosad A substance-ee theory of laryngeal features
12. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism What is substance-ee phonology? Why substance-ee phonology? The workings of the theory Towards a modular phonology A modular approach should involve some domain-specificity An uneasy position for classic generative phonology because of the Jakobsonian legacy of substantive markedness and universal features (Jakobson, Fant, and Halle 1951; Chomsky and Halle 1968) Contrast Fudge (1967); Foley (1977): generative phonology is wrong because it is ‘transformational phonetics’ Burton-Roberts (2000): phonology is not specifically linguistic in the generative sense, because it is so bound to substance Pavel Iosad A substance-ee theory of laryngeal features
13. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism What is substance-ee phonology? Why substance-ee phonology? The workings of the theory The importance of representations (even in OT) A modular theory is more restrictive than a fully parallel one In principle, OT can be done in a modular way (van Oostendorp 2007; Bermúdez-Otero 2012) This requires serious discipline in formulating constraints But constraints are always constraints on representations If phonology is a module, an aspect of its encapsulation should be the existence of a dedicated universe of discourse So phonetic substance should not come into it: non-trivial representational theory Answering the charge of Burton-Roberts (2000): if the phonological alphabet is not substance-bound, we can have a linguistic phonological module Pavel Iosad A substance-ee theory of laryngeal features
14. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism What is substance-ee phonology? Why substance-ee phonology? The workings of the theory How it works I The contrastivist hypothesis: as far as possible, phonology makes use only of features allowed in the lexicon (Dresher 2009; Hall 2007) Substance-ee representations Features are emergent and language-specific No a priori connection to substance Phonological patterns are the main evidence Non-trivial but constrained phonetics-phonology interface: the phonological analysis does not make simplistic predictions about how things should be pronounced Geometric approach: the Parallel Structures Model (Morén 2003, 2006, 2007; Krämer 2009; Youssef 2010; Iosad 2012c; Uffmann 2013) Tier structure: recursion of tiers Pavel Iosad A substance-ee theory of laryngeal features
15. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism What is substance-ee phonology? Why substance-ee phonology? The workings of the theory How it works II Privative (unary) features: no reference to minus values Structural size defines markedness relations without stipulation (contrast de Lacy 2006; Nevins 2010) Ternarity and the contrastive hierarchy Unlike other versions of the PSM (and other privative approaches), I allow a contrast between a bare node and the absence of a node (example to come) So ⟨×⟩ is not the same as ⟨×, C-lar⟩ Tier specification comes om the contrastive hierarchy à la Dresher (2009) Potential for ternary contrasts (Inkelas 1994; Krämer 2000; Strycharczuk 2012) Not a ee-for-all: since tier structure also defines markedness relationships and feature interaction, this is not (necessarily) a notational variant of binary features Pavel Iosad A substance-ee theory of laryngeal features
16. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Outline . . . 1 Substance-ee phonology: what, why, and how . . . 2 Celtic laryngeal phonology vs. laryngeal realism . . . 3 Substance-ee laryngeal realism Pavel Iosad A substance-ee theory of laryngeal features
17. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism An introduction to laryngeal phonology Trying to analyse laryngeal features in a theory like the present one, we face several choices Binary or ternary contrast? Specification of the less marked member: no node or bare node? What is the feature? . Proposal . . . Evidence om the Celtic languages shows that: Laryngeally unspecified obstruents are rarer than oen thought The choice of the marked value has no necessary connection to phonetics Pavel Iosad A substance-ee theory of laryngeal features
18. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Celtic ‘H languages’: phonetics Welsh, Scottish Gaelic and (most dialects of ) Irish have an obstruent system similar to that found in many Germanic languages Aspiration (even including preaspiration) of ‘voiceless’ stops See Ball (1984); Ball and Williams (2001); Morris (2010) for Welsh; Ladefoged et al. (1998); Nance and Stewart-Smith (forthcoming) for Scottish Gaelic; Ní Chasaide (1986); Hickey (2011); Ó Raghallaigh (2013) for Irish As in English, German, Icelandic, Swedish… ‘Voiced’ stops are not really voiced… Completely voiceless, short-lag VOT (Scottish Gaelic, also Icelandic, Danish) Incompletely voiced, especially in non-voiced contexts (Welsh, Irish, also English, German) Pavel Iosad A substance-ee theory of laryngeal features
19. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Celtic ‘H languages’: phonology Phonologically, all these languages seem to assign a feature to ‘fortis’ stops For instance, [h] in suffixes has a devoicing effect ⑴ Welsh a. [ˈteːɡ] teg ‘fair’ b. [ˈtɛkaχ] tecach ‘fairer’ ⑵ Cois Fharraige Irish (see especially Ó Buachalla 1985) a. [Lʲaːɡ] leag ‘put!’ b. [Lʲaːkə] leacfaidh ‘will put’ Pavel Iosad A substance-ee theory of laryngeal features
20. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Celtic ‘H languages’: phonology cont’d This makes sense if [h] is just C-laryngeal[spread glottis] and the voiceless stops are the result of coalescence ⑶ . . h2 . C-lar . [SG] . b1 . C-pl . [lab] . C-man . [cl] . C-lar . p1,2 . C-lar . [SG] . C-pl . [lab] . C-man . [cl] . ⇒ Pavel Iosad A substance-ee theory of laryngeal features
21. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Breton: a Celtic ‘L language’? Breton laryngeal phonetics is very different (Essentially like French) Short-lag VOT vs. prevoiced stops (e. g. Bothorel 1982) Phonologically, however, it provides robust evidence for the markedness of voiceless stops ⑷ Devoicing suffixes in Bothoa Breton (Humphreys 1995) a. [fɛbˈliːʒən] feblijenn ‘weakness’ b. [ˈfæːb̥] fepl ‘weak’ c. [ˈfæpɒh] feploc’h ‘weaker’ Analysis: Breton has a C-laryngeal[voiceless] feature Structurally, it is like Welsh and the rest Pavel Iosad A substance-ee theory of laryngeal features
22. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Ternarity in Breton I Moreover, Breton provides robust phonological evidence for ternary contrasts Word-final obstruents are fully delaryngealized: no C-lar specification Voiced obstruents are C-laryngeal Evidence: spreading of C-laryngeal[voiceless] across a word boundary blocked because there is no C-laryngeal node Pavel Iosad A substance-ee theory of laryngeal features
23. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Ternarity in Breton II ⑸ [ˌdɛnd̥ ˈhiːr] ‘long tooth’ (*[ˌdɛnt ˈhiːr]) . . d̥. C-man . [cl] . C-pl . [cor] . . dɛn . h . C-lar . [vcl] . iːr But C-lar[vcl] can spread if a C-lar node is provided When there is no le context, rightwards coalescence of the C-lar node produces voicing When there is a le context, C-lar[vcl] spreads lewards Pavel Iosad A substance-ee theory of laryngeal features
24. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Ternarity in Breton III ⑹ [ˌiːlis ˈkoːz̥] ‘old church’ (*[ˌiːliz ˈɡoːz̥]) . . k . C-lar . [vcl] . C-man . [cl] . C-lar . s . C-pl . [cor] . ili .oːz̥ The C-lar is provided by the initial consonant mutation context Analysis explains why the same context can trigger both voicing and devoicing (Krämer 2000) Word-final delaryngealization consistent with final devoicing and pre-sonorant voicing (e. g. Colina 2009, though see Strycharczuk 2012 for numerous caveats) Pavel Iosad A substance-ee theory of laryngeal features
25. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Laryngeal phonology in Celtic: summary Most Celtic languages show evidence of a contrast between C-laryngeal[‘fortis’] and C-laryngeal obstruents, irrespective of the phonetic realization Laryngeally unspecified obstruents appear in tightly circumscribed contexts The lack of laryngeal specification has both phonetic and phonological consequences Phonetic surface underspecification (Keating 1988) Inactivity on the laryngeal tier Pavel Iosad A substance-ee theory of laryngeal features
26. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Laryngeal realism: the basics The phonological analysis above has a lot in common with ‘laryngeal realism’ (Iverson and Salmons 1995, 1999, 2003a, 2003b, 2007; Honeybone 2005, 2012; Jessen and Ringen 2002; Petrova et al. 2006; Helgason and Ringen 2008; Beckman et al. 2011; Beckman, Jessen, and Ringen, forthcoming; Jansen 2004, 2007) ‘L languages’ like French and Hungarian: full voicing of lenis stops, phonological activity of [voice]/L ‘H languages’ like English and German: aspiration in fortis stops, passive voicing in lenis stops om surface underspecification, phonological activity of [spread glottis]/H Pavel Iosad A substance-ee theory of laryngeal features
27. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Laryngeal phonology in Celtic Laryngeal realism Laryngeal realism and substance-ee phonology In substance-ee phonology, the phonological part of the laryngeal realism argument is valid Phonological patterns are the evidence for featural representations The tight connection between the type of laryngeal contrast and phonetics is invalid In the remainder of this talk I argue that laryngeal realism faces problems precisely when it abandons purely phonological argumentation Pavel Iosad A substance-ee theory of laryngeal features
28. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Outline . . . 1 Substance-ee phonology: what, why, and how . . . 2 Celtic laryngeal phonology vs. laryngeal realism . . . 3 Substance-ee laryngeal realism Pavel Iosad A substance-ee theory of laryngeal features
29. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Some problems with laryngeal realism Laryngeal realism assumes a tight, two-way connection between voicing and phonological specification Full prevoicing ⇔ [voice] specification Partial voicing ⇔ laryngeal underspecification This is problematic in several respects, as we shall see shortly ‘Unspecified’ obstruents show controlled implementation even in ‘H’ languages I argue that the substance-ee approach solves both problems Pavel Iosad A substance-ee theory of laryngeal features
30. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Redundant specifications Some languages, notably Swedish (Helgason and Ringen 2008; Beckman et al. 2011), show a contrast between aspirated and fully voiced stops If full voicing equals phonological specification, we have a redundant system: H/L instead of H/∅ or ∅/L Swedish voiced stops are inert phonologically (Ringen and Helgason 2004) Substance-ee solution: Swedish contrasts C-lar[SG] and C-lar, and the phonology does not care about the phonetics of bare C-lar Still, bare C-lar is a specification associated with articulatory instructions These instructions happen to be conventionalized in Swedish as full voicing Pavel Iosad A substance-ee theory of laryngeal features
31. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Lack of passive voicing In languages like Icelandic, Danish, and Scottish Gaelic, the lenis stops are actively devoiced Not compatible with lack of specification Substance-ee solution: they contrast C-lar[SG] with C-lar, with a conventionalized implementation of C-lar Cues other than voicing (duration, pitch…) Pavel Iosad A substance-ee theory of laryngeal features
32. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again But what about English? Variable voicing in languages like English has been taken as evidence for laryngeal underspecification The default assumption in the present proposal is a C-lar specification (because of the contrastive hierarchy) The realization of lenis stops is clearly conventional (e. g. Docherty 1992; Scobbie 2006) Variable closure voicing does not mean lack of control (Westbury 1983; Westbury and Keating 1986; Kingston and Diehl 1994; Kingston, Lahiri, and Diehl 2009): duration, pitch, expansion of the supraglottal cavity Same argument for Welsh Pavel Iosad A substance-ee theory of laryngeal features
33. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Why do cues recur? Laryngeal realism expresses a real insight into how laryngeal contrasts tend to be implemented Despite the arbitrary and conventional nature of the cues, they do tend to recur a lot Isn’t that an explanandum? Answer: it is, but the explanation is not the job of phonology Pavel Iosad A substance-ee theory of laryngeal features
34. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Looking for extraphonological explanations Kingston and Diehl (1994; 1995); Kingston et al. (2008, et passim) propose the concept of intermediate perceptual properties (IPPs), which integrate different cues into unified percepts IPPs mediate between diverse acoustic cues (closure voicing, pitch movements, durational properties, burst strength etc.) and what appear to be monolithic phonological specifications Crucially, Kingston et al. (2008) argue that IPPs are not specifically linguistic, but are part of the general perceptual system If IPPs come for ee, it is natural for learners to use them in acquisition, explaining the recurrence of cues Pavel Iosad A substance-ee theory of laryngeal features
35. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Diversity of cues: summary Even though the realization of contrasts in substance-ee phonology is in principle conventional and arbitrary, the availability of IPPs biases learners towards reusing a particular (relatively small) set of cues See Kirby (2010) on the dynamics of the relationships between those cues The phonetic realization of the laryngeal contrast has nothing to do with the phonological markedness relationships either conceptually or empirically Where this relationship does exist, it is due to diachronically actuated biases rather than phonological principles Pavel Iosad A substance-ee theory of laryngeal features
36. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Conclusion The geometric theory of laryngeal contrast provided here salvages the important phonological insights of laryngeal realism regarding the relative markedness of obstruent classes At the same time it provides an empirically better theory of ternary contrasts and surface underspecification Substance-ee phonology gives just enough leeway to phonetic implementation to remain empirically adequate A lot of what used to be ‘phonology’ is now outside of it A better theory of where phonological computation stands in the study of language promises a better relationship with more empirical work Pavel Iosad A substance-ee theory of laryngeal features
37. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . .. . . . .. . . . . .. . . . .. . . . . .. . . . .. . . . .. . Substance-ee phonology: what, why, and how Celtic laryngeal phonology vs. laryngeal realism Substance-ee laryngeal realism Problems with laryngeal realism Substance-ee solutions Coupling phonetics with phonology again Conclusion The geometric theory of laryngeal contrast provided here salvages the important phonological insights of laryngeal realism regarding the relative markedness of obstruent classes At the same time it provides an empirically better theory of ternary contrasts and surface underspecification Substance-ee phonology gives just enough leeway to phonetic implementation to remain empirically adequate A lot of what used to be ‘phonology’ is now outside of it A better theory of where phonological computation stands in the study of language promises a better relationship with more empirical work Thank you! Tapadh leibh! Pavel Iosad A substance-ee theory of laryngeal features