LINGUIST List 16.1790|
Mon Jun 06 2005
Review: Language Acquisition/Phonology: Ota (2003)
Editor for this issue: Naomi Ogasawara
What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in. If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Sheila Dooley at collberglinguistlist.org.
The Development of Prosodic Structure in Early Words
Message 1: The Development of Prosodic Structure in Early Words
From: Geoffrey Nathan <geoffnathanwayne.edu>
Subject: The Development of Prosodic Structure in Early Words
AUTHOR: Ota, Mitsuhiko
TITLE: The Development of Prosodic Structure in Early Words
SUBTITLE: Continuity, divergence and change
SERIES: Language Acquisition and Language Disorders 34
PUBLISHER: John Benjamins
Announced at http://linguistlist.org/issues/15/15-371.html
Geoffrey S. Nathan, Department of English, Wayne State University
This is an Optimality-Theory (OT)-based study of the early acquisition
of some part of the prosodic structure of Japanese by native
speakers. It deals primarily with how young Japanese speakers cope
with mora and syllable structure and how they come to follow the
constraints that Japanese imposes on these structures. It also deals
with the nature of children's 'underlying representations', whether
children's phonologies contain different mechanisms and ontologies
from those of adults and how child phonology might evolve towards
adult phonology. The book is a revised version of the author's
Georgetown University doctoral dissertation.
Chapter 1 introduces the subject, the relation of phonological theory
to language acquisition, and the fundamental question of whether
children's phonologies are built of different 'stuff' from those of adults,
or merely adult phonologies rearranged in various ways (say with the
same constraints in different rankings). In addition it raises the
question of whether children have the same lexical representations as
adults, merely modified through interfering constraints, or whether
they have distinct 'underlying' representations that are closer to the
modified forms they actually produce.
Chapter 2 introduces basic phonological constructs such as the
prosodic hierarchy (Prosodic Word, Foot, Syllable and Mora), foot
types, mora types and so on, and provides a basic introduction to the
essential aspects of Japanese phonology, including accent, feet and
moraic structure. It also provides a basic introduction to Optimality
Theory, then summarizes some early research on children's
acquisition of moraic structure in several languages, including English,
Dutch and Spanish. Finally it deals with the question of whether
children are setting parameters or reranking constraints as they
grapple with the more difficult aspects of their native phonologies.
Chapter 3 provided the methods used, specifically tape and video
recordings of natural interactions between three monolingual
Japanese children (aged 1;0-1;5 at the beginning of the study and 2;0-
2;6 at the end). Data were transcribed phonetically, and some
utterances were analyzed with X-waves.
Chapter 4 deals with the representation of syllable-internal structure.
The children studied showed evidence that they were aware of
syllable weight, and that it is determined solely by the presence vs.
absence of branching rhymes (i.e. that onsets have no effect on
weight). The evidence for this includes the fact that the children often
produced words containing heavy syllables with some kind of weight,
but not always the kind found in the target. Thus CVC syllables were
replaced with CV: syllables (and never vice versa), and CV1V2
syllables were replaced with CV: as well. Secondly, 'superheavy'
syllables were frequently simplified to merely heavy (here the nature
of the simplification proceeded in both directions: CVVC --> CVC and
CVV). Finally, children seemed to be providing evidence for Zec's
(1988) argument that the sonority hierarchy is active within the
universals of coda consonant selection (if a language permits
obstruents as the second mora in a syllable it permits sonorants and
Chapter 5 deals with the development of syllable-internal structure.
Using the Optimality Theory framework, Ota argues that the gradual
acquisition of syllable codas can be modelled through the assumption
that in the 'initial state' markedness constraints outrank faithfulness
constraints, and that various faithfulness constraints are reranked
above the relevant markedness constraints as various kinds of codas
become licit. For example, the children, in stage one (just over a year)
had no codas at all, pronouncing /mamma/ 'food' as [mama]
and /wanwan/ 'doggie' as [wawa]. Later, in stage 2, nasals were
permitted in codas but obstruents were forbidden. By 'fiddling with' the
Markedness constraints on what kinds of segments can serve in a
mora slot the author can derive the appropriate ordering of
acquisition. In some cases she is forced to wrestle with the problem of
negative evidence for reordering, and ends up arguing that sometimes
reranking of three different constraints can be achieved through
changes in ranking of two different pairings, coupled with an
assumption of Lexicon Optimization. In addition, there is evidence that
children differentiate between word-internal syllable codas (which
exhibit compensatory lengthening when a coda-consonant is 'deleted')
and word-final codas, which drop without remedy.
Chapter 6 covers word-level effects. Adult Japanese has no minimum
size restrictions on (underived) lexical items, and common words such
as 'hand' /te/ and 'tree' /ki/ illustrate this. The children, on the other
hand do seem to have such restrictions, lengthening such words and
even lengthening truncations of longer words with 'underlying' short
vowels (such as /banana/ becoming [ba:].) Such a process is active in
adult derived phonology (see, for example Ito 1990 and Mester 1990),
but in such an obscure way that it is unlikely children have had much
opportunity to observe its operation. It is presumably an instantiation
of 'the emergence of the unmarked' in child speech (in addition to its
occurrence per se within the phonology of Japanese).
Adult Japanese words also have no maximum size restrictions, but the
children apparently did have one--they generally had no words larger
than a well-formed binary foot. Thus they had words of shape CV:
([ba:] from /oba:tyan/ 'grandma') and CV(:)CV ([ji:da]
from /zido:sya/ 'car') and no larger. Thus words can be no smaller
than bimoraic, but no larger than bisyllabic. Again, this is a restriction
the children are imposing on the language, rather than vice versa.
Chapter 7 deals with the developmental path shown by word-sized
restrictions as children are increasingly able to produce words longer
than a single foot. Again, the developmental path can be modelled by
the reshuffling of constraints, with markedness constraints moving
rightwards (or faithfulness constraints of various kinds moving
Chapter 8, Conclusions, summarizes the analyses throughout the
book. The basic points are that children have the same set of
constraints as adults, but ranked in different ways, and that children's
underlying representations are the same as the adult surface form. As
has been argued by many OT theorists, children begin with
markedness constraints reigning supreme, but faithfulness constraints
gradually assert dominance as children learn how to pronounce target
sequences despite the 'inertia' of their natural inclinations. To some
extent this causes problems, because a standard assumption of most
flavors of generative phonology (OT included) is that predictable
information is not stored lexically. Since the mere presence of a
consonant in a coda defines a syllable as heavy, a separate moraic
marker is not needed in addition to the presence of the consonant per
se. However, these children appear to be applying the (formerly
called) process of compensatory lengthening, 'replacing' codas with
vowel length. Within the OT framework this can only be accounted for
if the mora attached to the coda is actually specified lexically, a
redundant specification. Ota argues that Lexicon Optimization will
permit this kind of specification, in the absence of any notion of
As mentioned above, this book is a somewhat revised PhD
dissertation. This is both its strength and its weakness. Its strength is
that it is a very detailed and careful study of a small topic, exactly what
we want a dissertation to be. Its weakness is that it is structured like a
dissertation, with each chapter consisting of exactly the same,
somewhat ponderous structure: 'first I'll tell you what I'm going to
show, then I'll show it to you, then I'll tell you what I've shown you.'
Within each section there is the same, somewhat repetitive rhetorical
style, so that the reader is left saying 'come on! just say it and get it
Despite the somewhat ponderous style, however, the data, and
analyses in this book are very interesting, dealing in great detail with a
topic that has not been studied much--the fine-grained structure of the
acquisition of syllable and word structure in a mora-timed language.
Unfortunately the author chose not to present many real examples of
the kinds of things the children did, presenting the data in statistical
tables instead (a list of examples over time, similar to that found, for
example, in the classic work by Smith (Smith, 1973) would have been
very useful). On the other hand, the facts that Ota uncovered are grist
for many theoretical mills. Clearly children know about aspects of
language that they have not been exposed to (the classic 'emergence
of the unmarked' effects, such as the absolute size restrictions the
children exhibited at an early stage), which poses problems for those
developing usage-based models of language acquisition (Bybee,
2001). Similarly, children replace coda consonants with lengthened
vowels (essentially doing compensatory lengthening 'on the fly'), for
which there is no evidence in the ambient language data, but which
recapitulates well-known historical processes. Such behavior,
emphasized in Natural Phonology (particularly in Stampe, 1969;
Donegan & Stampe, 1979) has, in some cases been 'replicated' with
various mechanisms of OT, but still needs to be accounted for in any
phonological theory that deals with the mysterious knowledge children
seem to have of how their language is 'supposed to' work, even when
it doesn't. As others have occasionally pointed out, the use of the OT
formalism, while now virtually an 'industry standard', provides no more
explanatory mechanisms than other notations that might have been
used in earlier periods. The use of Tesar and Smolensky's (1998)
theory of learnability is the closest thing to an actual explanatory
account the book contains.
Finally there is a more general problem with all works of this kind. It
relies heavily on the notational conventions of OT. While OT is
currently the dominant paradigm, its notational conventions require
considerable effort to learn. This is not a problem as long as we can
all read them, but for a book to have any shelf life the analyses need
to be presented in a notational manner that can be read by future
generations. How many of our students can read the Main Stress Rule
in the Sound Pattern of English (Chomsky & Halle, 1968)? Similar
issues arise with some of the grammars of endangered languages
written during various theoretical episodes in our field's history. How
many future generations can read, and get something useful from,
syntactic descriptions written in Revised Extended Standard Theory
notation (such as it was), or some of the more exotic structuralist
notations, such as the original Newman article on Yawelmani
(Newman 1944). The problem is analogous to that of electronic
storage formats. How many of us own software that can read
Wordstar files? How much longer will CD-ROM drives be enough of a
standard that we can read the backup files stored on them?
Bybee, J. L. 2001. Phonology and Language Use. Cambridge and
New York: Cambridge University Press.
Chomsky, N., & Halle, M. 1968. The Sound Pattern of English. New
York: Harper and Row.
Donegan, P. J., & Stampe, D. 1979. The study of Natural Phonology.
In Dan Dinnsen (Ed.), Current Approaches to Phonological Theory..
Bloomington: Indiana University Press.
Ito, Junko. 1990. Prosodic minimality in Japanese. in M. Ziolkowski, M.
Noske and K. Deaton (Eds) Papers from the 26th Meeting of the
Chicago Linguistic Society. CLS: Chicago.
Mester, Armin. 1990. Patterns of truncation. Linguistic Inquiry 21: 478-
Newman, Stanley. 1944. Yokuts Language of California. New York:
Viking Fund Publications in Anthropology 2.
Prince, Alan and Paul Smolensky. 1994  Optimality Theory:
Constraint Interaction in generative grammar. Slightly revised version
as ROA 537-0802 at roa.rutgers.edu
Smith, N. V. 1973. The Acquisition of Phonology: A case study.
Cambridge: Cambridge University Press.
Stampe, D. 1969. The acquisition of phonetic representation. In
Papers from the fifth regional meeting of the Chicago Linguistic
Society (pp.a443-454). Chicago: Chicago Linguistic Society.
Tesar, Bruce & Paul Smolensky. 1998.  Learnability in
Optimality Theory. Linguistic Inquiry 29: 229-68. Reprinted in John J.
McCarthy. 2004. Optimality Theory in Phonology. Malden and Oxford:
Blackwell. pp. 118-140.
Zec, Draga. 1988. Sonority constraints on prosodic structure. Doctoral
Dissertation, Stanford University.
Zec, Draga. 1995. Sonority constraints on syllable structure.
Phonology 12, 85-129.
ABOUT THE REVIEWER
Geoffrey S. Nathan is Associate Professor of Linguistics in the English
Department at Wayne State University. His research interests include
phonological theory, phonetics, second language acquisition and the
history of linguistics. He is also Faculty Liaison and Security Policy
Coordinator for Computing and Information Technology at Wayne.
Respond to list|Read more issues|LINGUIST home page|Top of issue
Please report any bad links or misclassified data
LINGUIST Homepage | Read
LINGUIST | Contact us
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.