Tobias Scheer CNRS 7320, Université de Nice - Sophia Antipolis [email protected]

DGFS 7-9 March 2012 Frankfurt am Main

this handout and some of the references quoted at www.unice.fr/scheer/

HOW PHONOLOGICAL IS INTONATION? (1)

cocktail version of the talk a. goal to examine the relationship of intonation (sentence stress) with phonology in general, and with recursion in particular. Definition: "phonology" = the computational system that assigns phonological structure, i.e. which transforms inputs into outputs. b. conclusion 1. intonation lies outside of phonology 2. recursive phonological structure is the result of analyses that try to account for intonation in the wrong module. c. alternative (more or less following Wagner 2005) 1. intonation is a purely syntactic phenomenon. Syntactic information (numeration, information structure, semantics etc.) alone defines the locus of intonation. ==> phonological computation does not participate in the definition of intonation. 2. phonology (or rather: phonetics?) interprets the carrier of intonation information, which is a regular Vocabulary Item that has come into being through regular spell-out. [this talk is based on Scheer 2011:§800]

1. Some undisputed facts (2)

Intonation is governed by syntactic structure a. At least since Bresnan (1971), it is established that intonation (also called sentence or phrasal stress) directly depends on syntactic structure. The topic is covered by a rich syntactic literature, including Berman & Szamosi (1972), Cinque (1993), Kahnemuyipour (2009) and Adger (2007). b. syntax-sensitive intonation: classical example from Bresnan (1971) A) Helen left directions for George to follow. B) Helen left directions for George to follow. 1. A) means that Helen has left some directions that George should follow 2. B) is an invitation for George to follow Helen c. Since both sentences are phonologically identical but have contrasting syntactic structure, the different intonation must be a consequence of the latter: under A) follow is transitive and belongs to a relative clause whose head is directions, while under B) it is intransitive and complements directions.

-2(2)

Intonation is governed by syntactic structure d. The causal relationship between syntax and phonology has also been thought of in the opposite direction: SzendrEi (2001, 2003, 2004) for example argues that syntactic properties such as focus and climbing can be controlled by intonation. Intonation is thus a phenomenon that challenges Zwicky & Pullum's (1986) original (strong) version of phonology-free syntax (which is probably more accurately referred to as melody-free syntax).

(3)

there is no recursion in phonology a. this is a pervasive and largely undisputed empirical generalisation which has been recognized for long as a major watershed line that separates morpho-syntax and phonology (and semantics). b. E.g. Pinker & Jackendoff (2005a,b), Neeleman & van de Koot (2006) c. Hauser et al. (2002) make recursion (Merge) even the only content of FLN, i.e. which cannot be derived from animal-shared FLB through evolution.

2. Phonological vs. syntactic trees? (4)

why there is no recursion in phonology: because there is no concatenation a. the to-date unchallenged inverted T model of the architecture of grammar (Chomsky 1965:15ff) establishes the idea that 1. morpho-syntax has the privilege of concatenation 2. phonology and semantics only interpret b. hence phonology and semantics are not equipped for gluing pieces together. c. in the minimalist environment, concatenation is the result of Merge. This operation is thus available in morpho-syntax, but not in phonology and semantics. d. 1. morpho-syntactic recursion is due to Merge (and to nothing else) 2. Merge is the concatenative device 3. ==> no concatenation, no recursion ==> there is no recursion in phonology because nothing is concatenated. Scheer (2004:xliv, 2011:§42).

(5)

what about phonological trees? a. phonological theories have always relied on tree-building devices, at least since the advent of autosegmental structure. b. feature geometric trees are lexically specified: no online construction c. but syllabic and prosodic arborescence is assumed to be the result of online treebuilding activity d. syllabification algorithms build arboreal syllable structure on the basis of the segmental properties of a lexically unsyllabified linear string. e. BUT, again: phonological trees do not involve any concatenation of pieces (they are built on a pre-existing linear string): this is what makes them different from morpho-syntactic trees. f. ==> phonological and morpho-syntactic trees are not the same thing. g. thus if any, the phonological tree-building device is different from morpho-syntactic Merge h. accommodating distinct Mergem-synt and Mergephon in grammatical theory of course ruins the minimalist ambition, which counts on only one universal piece-gluing (and hence tree-building) device.

-3(5)

what about phonological trees? i. [there is also a FLAT alternative for syllable theory where the function of trees is taken over by lateral relations: Government Phonology in general, and so-called CVCV in particular (deforestation, lateralization of structure and causallity). Scheer (2004:§165, 2011:§42)

(6)

more reason to believe that there is no tree-building device in phonology a. Neeleman & Van de Koot (2006) show that trees of whatever kind have certain formal properties that make predictions on the type of phenomena that should be found in a tree-bearing environment. b. these include 1. projection 2. long-distance dependencies and 3. recursion. c. Neeleman & van de Koot (2006) demonstrate that phonological phenomena do not display any of these properties. d. they therefore conclude that the presence of trees in phonology overgenerates: arboreal structure predicts things that are absent from the record.

3. Nested prosodic structure in the analysis of intonation (7)

recursion in phonology, due to the analysis of intonation a. as far as I can see, Ladd (1986) was the first voice to argue that intonation is recursive, in the sense that it requires recursive prosodic structure. b. definition of recursion a structure is recursive iff there is a node that is dominated by another node of the same kind. c. Ladd (1986) works with two prosodic constituents, the Major Phrase and the Tone Group (TG). He aims at showing that intonation cannot be adequately described unless an MP may dominate other MPs, and a TG other TGs. d. nested prosodic structure was ruled out by the original version of Selkirk's Strict Layer Hypothesis, and this is what Ladd (1986) stands up against. e. under this pressure (among others), Selkirk (1996) abandons the ban on recursive prosodic structure: in the new constraint-based formulation, the non-recursion of prosodic structure is demoted to a violable constraint f. since then, numerous analyses have taken advantage of this option (among many others, Booij 1996, Peperkamp 1997, Truckenbrodt 1999).

(8)

what's phonology got to do with it? a. the bare existence of recursive prosodic structure in phonology does not mean that phonology itself is recursive as long as the structure in question is built by mechanisms that are foreign to phonological computation. b. this is indeed the case: mapping, i.e. the translation of morpho-syntactic into prosodic structure, is done outside of the phonology, and with no contribution of phonological computation. ==> more about that below. c. remember the definition of phonology: phonological computation.

-4-

4. Confusion between recursive phenomena and recursive analyses (9)

phenomena vs. analysis a. recursion in morphology and syntax exists on the grounds of a pre-analytical and pre-theoretical definition of recursion. By contrast, analysis is the only reason why recursion is said to exist in phonology. There is no pre-analytical or pre-theoretical evidence: ==> there is no recursive phenomenon, there are only recursive analyses of nonrecursive phenomena. ==> in morpho-syntax, there are recursive analyses of recursive phenomena.. b. recursive phenomenon in syntax: embedded clauses Peter thinks [that John says [that Amy believes [that…]]] c. recursive phenomenon in morphology: repeated affixes Czech iterative -áv dPlat "to do" dPl-áv-at "to do repeatedly/often" dPl-áv-áv-at "to do even more often" dPl-áv-áv-áv-…-at "to do really really often" French re- prefixation (about the same in English) faire "to do" re-faire "to do again" re-re-faire "to do with two repetitions" re-re-re-faire "to do with three repetitions" re-re-re-re-…-faire "to do with n repetitions" d. definition (pre-analytical and pre-theoretical) you keep doing the same thing until extra-grammatical restrictions such as memory etc. stop you. 1. production of grammatically unbounded embedding: grammar happily generates and tolerates an infinite number of embedded clauses (or phrases) 2. the limits on recursive structure in actual production are imposed by performance (factors such as memory), not by competence. 3. that is, speakers will get confused upon the third of fourth level of embedding.

(10) there is no recursive phenomenon in phonology a. nothing equivalent has ever been reported in phonology (including intonation). b. this empirical situation is the reason why the absence of recursion is firmly established as a major property that sets phonology (and semantics) apart from morpho-syntax (see above). c. arguing for an analysis of a phenomenon that recurs to recursive prosodic constituents does not make this phenomenon recursive. d. the same goes for analyses of other phonological phenomena that use recursive constructions. Van der Hulst (2010) has gathered a number of this kind of analyses regarding for example the internal structure of segments (melodic organisation).

-5-

5. Arguments (11) argument 1 absence of recursion in phonology – except for intonation?? a. intonation is the only area in phonology where analysts regularly propose recursive structure b. hence something must be wrong: in a modular environment, either a given computational system is able to produce recursive structure, or it is not. c. in case it is, we expect recursive structure everywhere in its realm of competence. d. if thus recursion is restricted to intonation, the conclusion is that there must be two distinct computational systems: 1. one, call it A, that is unable to produce recursion (phonology minus intonation) 2. another, call it B, which is able to handle recursion (intonation) e. or, put differently, 1. if we have good reason to believe that there is no recursion in phonology and 2. if we know why there is no recursion in phonology (because nothing is concatenated) the conclusion must be that whatever is recursive and looks phonological isn't phonology: ==> computational system B, whatever it is, lies outside of phonology. (12) argument 2 can the intonation of a sentence be predicted without knowing which words will fill in the slots ? a. hence knowing all about - the syntax of a sentence - its semantics - its information structure ==> can its intonation be predicted? b. in case it can, it is impossible that phonology has got anything to do with intonation: intonation must be calculated before lexical (vocabulary) insertion (VI), and hence before phonological computation has a chance to do anything at all. c. the answer appears to be positive: intonation can be predicted before VI: Féry & Ishihara (2009) the particular words and their phonological properties that will ultimately instantiate the syntactic structure are irrelevant for the computation of intonation. (13) argument 3 intonation and phonology appear to live in waterproof worlds a. were intonation determined by phonological computation, it should interact with other aspects of this computation. ==> this does not seem to be the case. b. has intonation ever been observed to impact any non-intonational phonological property (segmental, word-stress, tone, syllable structure etc.)? c. has any non-intonational phonological property ever been observed to impact intonation?

-6(13) argument 3 intonation and phonology appear to live in waterproof worlds d. note the contrast with real phonological phenomena: 1. word-stress is sometimes sensitive to syllable structure (weight by position), or to major categories (weight conferred by coda sonorants, but not by coda obstruents) (e.g. Szigetvári & Scheer 2005). 2. tone is often sensitive to syllable weight (contour tones only on heavy syllables). 3. etc. (14) argument 4 prosodic structure may be recursive – phonology is not a. let us assume the analyses using recursive prosodic constituency are correct: ==> this does not mean that phonology is recursive. b. on standard assumptions, prosodic constituency is not created by phonological computation: it is the result of a mapping mechanism that is located in modular no man's land, and the mapping computation is carried out by a specific interface module. E.g. Nespor & Vogel (1986), Jackendoff (2002). c. the story may be a little different since the Prosodic Hierarchy has been OTed: 1. constraint-based mapping has replaced rule-based mapping ==> ALIGN (and WRAP) 2. constraint-based mapping shifts mapping INTO the phonology: ALIGN constraints are interspersed with regular phonological constraints. ==> a harsh violation of modularity, since ALIGN also computes morphosyntactic information ("align the prosodic word with XP"), which however phonology is unable to parse on modular assumptions (domain specificity). 3. but even with constraint-based mapping, the prosodic hierarchy must somehow exist before phonology, i.e. ALIGN, starts to work. discussion: Scheer (2011:§457), Scheer (forth a,b). d. what that means is that everything is fine with recursive prosodic constituency as long as it is understood that prosodic constituents are NOT phonology: 1. like other Vocabulary Items, they are created elsewhere and INSERTED into the string that phonological computation takes as an input. 2. like for other VI, phonological computation assigns a pronunciation to, say, Intonation Phrases (IP) 3. but phonological computation cannot/does not modify prosodic constituency: prosodic constituents are interpreted, but not created or manipulated.

-7-

6. Syntax defines, phonology (or phonetics) executes (15) why replicating recursive structure in phonology? a. it was noted in (14)d that the current mainstream, recursive prosodic structure, may be fine with the absence of intonation from phonology. b. a question, though, is why recursion should be replicated in phonology when it does not do any labour in phonological computation. c. it is undisputed that all recursive structure is ultimately syntactic: if prosodic constituents are recursive, they simply mirror a recursive syntactic structure in the phonology. This is what prosodic structure does anyway: to import morphosyntactic information into phonology. d. Cinque's (1993) and Wagner's (2005) work shows that intonation is much more syntactically bound than the phonological literature may suggest. e. Cinque (1993) derives the parameterisation of the Nuclear Stress Rule (which is responsible for English intonation in SPE and much subsequent work) from purely syntactic factors. As a consequence, the Nuclear Stress Rule and phonological treatments thereof (such as Halle & Vergnaud's 1987:263ff grid-based implementation that Cinque discusses specifically) are redundant and have to go: "if the effects of the Nuclear Stress Rule […] depend entirely on the direction in which depth of embedding develops, the rules become redundant. They merely recapitulate what follows from purely syntactic parameters. Hence they should be eliminated." Cinque (1993:244) (16) morphemic vs. boundary information Scheer (2011, forth a) a. syntactic structure may be spelled out as - morphemic information, or as - non-morphemic information: hash marks, omegas etc. (= boundary information) b. whatever the syntactic (or structural) identity of intonation, it is boundary information. c. as such it is transformed by spell-out into a regular vocabulary item that is inserted into the input string to phonological computation. d. ==> phonology (or rather: phonetics?) interprets the carrier of intonation information, but it does not participate in its definition, or insertion. e. this carrier comes into being through spell-out. The question is what it looks like: 1. redundant (and diacritic) prosodic structure that does not do any labour in phonology 2. a static Vocabulary Item hooks on syllables (or words) in order to mark their intonational properties: something like Pierrehumbert's (1980) tones? [Pierrehumbert 2000, Beckman & Pierrehumbert 1986, 1988 and Ladd 2000 provide overviews]

-8(17) conclusion a. intonation and its definition is absent from phonological computation. b. intonation "transits" in phonological representations as some Vocabulary Item c. 1000 dollar question: this is inconsistent with argument 3: intonation and regular phonology do not interact. If there is a VI that can be parsed by phonological computation, we expect it to be able to participate in this computation, i.e. the influence regular phonology and to be influenced by it. d. sidestepping phonology? is there a way some property (or item) can be defined in syntax and executed by phonetics, while being absent from phonology? References Adger, David 2007. Stress and Phasal Syntax. Linguistic Analysis 33: 238-266. Beckman, Mary & Janet Pierrehumbert 1986. Intonational Structure in Japanese and English. Phonology Yearbook 3: 15-70. Beckman, Mary & Janet Pierrehumbert 1988. Japanese Tone Structure. Cambridge, MA: MIT Press. Berman, A. & M. Szamosi 1972. Observations on sentential stress. Language 48: 304-325. Booij, Geert 1996. Cliticization as prosodic integration: The case of Dutch. The Linguistic Review 13: 219-242. Bresnan, Joan 1971. Sentence stress and syntactic transformations. Language 47: 257-281. Chomsky, Noam 1965. Aspects of the Theory of Syntax. Cambridge, Mass.: MIT Press. Cinque, Guglielmo 1993. A null theory of phrasal and compound stress. Linguistic Inquiry 24: 239-297. Féry, Caroline & Shinichiro Ishihara 2009. How Focus and Givenness Shapes Prosody. Information Structure from Different Perspectives, edited by Malte Zimmermann & Caroline Féry, 36-63. Oxford: OUP. Halle, Morris & Jean-Roger Vergnaud 1987. An Essay on Stress. Cambridge, Mass.: MIT Press. Hauser, Marc, Noam Chomsky & Tecumseh Fitch 2002. The faculty of language: what is it, who has it, and how did it evolve ? Science 298: 1569-1579. Hulst, Harry van der 2010. A note on recursion in phonology. Recursion and Human Language, edited by Harry van der Hulst, 301-341. Berlin: de Gruyter. Jackendoff, Ray 2002. Foundations of Language. Brain, Meaning, Grammar, Evolution. Oxford: Oxford University Press. Kahnemuyipour, Arsalan 2009. The Syntax of Sentential Stress. Oxford: OUP. Ladd, Robert 2000. Bruce, Pierrehumbert, and the Elements of Intonational Phonology. The interaction of constraints on prosodic phrasing. Prosody: Theory and Experiments, edited by Merle Horne, 37-50. Dordrecht: Kluwer. Neeleman, Ad & Hans van de Koot 2006. On syntactic and phonological representations. Lingua 116: 1524-1552. Nespor, Marina & Irene Vogel 1986. Prosodic Phonology. Dordrecht: Foris. Peperkamp, Sharon 1997. Prosodic Words. The Hague: Holland Academic Graphics. Pierrehumbert, Janet 1980. The phonology and phonetics of English intonation. Ph.D dissertation, MIT. Pierrehumbert, Janet 2000. Tonal Elements and their Alignment. The interaction of constraints on prosodic phrasing. Prosody: Theory and Experiments, edited by Merle

-9Horne, 11-36. Dordrecht: Kluwer. Pinker, Steven & Ray Jackendoff 2005a. The faculty of language: what's special about it ? Cognition 95: 201-236. Pinker, Steven & Ray Jackendoff 2005b. The nature of the language faculty and its implications for the evolution of language (Reply to Fitch, Hauser and Chomsky). Cognition 97: 211-225. Scheer, Tobias 2004. A Lateral Theory of Phonology. Vol.1: What is CVCV, and why should it be? Berlin: Mouton de Gruyter. Scheer, Tobias 2011. A Guide to Morphosyntax-Phonology Interface Theories. How ExtraPhonological Information is Treated in Phonology since Trubetzkoy's Grenzsignale. Berlin: Mouton de Gruyter. Scheer, Tobias forth a. Direct Interface and One-Channel Translation. A Non-Diacritic Theory of the Morphosyntax-Phonology Interface. Vol.2 of A Lateral Theory of phonology. Berlin: de Gruyter. Scheer, Tobias forth b. Chunk definition in phonology: prosodic constituency vs. phase structure. Modules and Interfaces, edited by Anna Bloch-Rozmej & Maria BlochTrojnar. Lublin: Institute of English of the Catholic University of Lublin [available at ling.auf.net/lingbuzz]. Selkirk, Elisabeth 1996. The prosodic structure of function words. Signal to syntax: bootstrapping from syntax to grammar in early acquisition, edited by James Morgan & Katherine Demuth, 187-213. Mahwah, NJ: Erlbaum. SzendrEi, Kriszta 2001. Focus and the Syntax-Phonology Interface. Ph.D dissertation, University College London. SzendrEi, Kriszta 2003. A stress-based approach to the syntax of Hungarian focus. The Linguistic Review 20: 37-78. SzendrEi, Kriszta 2004. A stress-based approach to climbing. Verb clusters. A study of Hungarian, German and Dutch, edited by Katalin É.Kiss & Henk van Riemsdijk, 205233. Amsterdam: Benjamins. Szigetvári, Péter & Tobias Scheer 2005. Unified representations for the syllable and stress. Phonology 22: 37-75. Truckenbrodt, Hubert 1999. On the Relation between Syntactic Phrases and Phonological Phrases. Linguistic Inquiry 30: 219-255. Wagner, Michael 2005. Prosody and Recursion. PhD. dissertation, MIT. Zwicky, Arnold & Geoffrey Pullum 1986. The Principle of Phonology-free Syntax: introductory remarks. Ohio State University Working Papers in Linguistics 32: 63-91.