LINGUIST List 14.1956

Fri Jul 18 2003

Review: Phonology/Morphology: Vig�rio, Marina (2003)

Editor for this issue: Madhavi Jammalamadaka <>

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 Simin Karimi at


  1. Gisela Collischonn, The Prosodic Word in European Portuguese

Message 1: The Prosodic Word in European Portuguese

Date: Thu, 17 Jul 2003 15:02:40 -0400 (EDT)
From: Gisela Collischonn <>
Subject: The Prosodic Word in European Portugese

Vig�rio, Marina (2003) The Prosodic Word in European Portuguese, Mouton de Gruyter, Interface Explorations 6.

Announced at

Gisela Collischonn, Universidade Federal do Rio Grande do Sul, Brazil


This book, a revised version of the author's doctoral thesis, is a
phonological investigation of prosodic structure at the level of the
prosodic word in European Portuguese (EP). An array of phonological
properties that identify the prosodic word in EP are presented and,
based on them, the prosodic structure related to affixes, clitics and
compound words is proposed. It provides empirical support for prosodic
theory, since it shows that the prosodic word has a role to play in
EP. It argues also for the pertinence of the division into lexical and
postlexical components in EP phonology. The thesis is divided into
eight chapters. The first four provide the theoretical and empirical
background; the following two chapters form the main part of the
book. Chapter 5 presents the argumentation for specific prosodic
structure associated to prefixes, suffixes, pronominal clitics and
other clitic words. Vig�rio takes phonological evidence to
support the view that prefixes, unlike suffixes, are adjoined to the
following prosodic word creating a recursive prosodic structure. The
same asymmetry is argued to explain the distribution of proclitics
vs. enclitics. Chapter 6 argues that two prosodic words that belong
to the same morphological or syntactic compound form a recursive
structure, the compound prosodic word, which is distinct from the
regular p-phrase. Chapter 7 provides empirical data about the
reduction of function words and discusses the generality of these
reduction processes and their status in grammar. Chapter 8 summarizes
the results obtained in the previous chapters and points to issues of
further research. The thesis comprises also two appendices (containing
collected data (Appendix I) and tables containing the results for the
empirical test of chapter 7), a list of references, and an index.
Critical evaluation This work is a contribution to the study of the interface between
phonology and morphosyntax, with special emphasis on the mapping of
prosodic words from morphosyntactic structure. It is also a
contribution to the description of EP since it displays a large amount
of data about segmental processes, as well as prominence and tonal
phenomena and phonotactic restrictions that had never been discussed
before or had not been discussed in relation to the issue of prosodic
word structure. The book displays empirical support for some of the ideas proposed in
Peperkamp (1997), Booij (1996) and others, about the prosodization of
clitics and prefixes. It does not propose any novel theoretical
mechanisms (except for the compound prosodic word), and its value lies
mainly in demonstrating how some recent proposals in prosodic
phonology (Peperkamp, 1997, Selkirk, 1995, Hayes, 1990) can be
successfully employed in analysing EP data, uncover unnoticed aspects
of the ponology of this language and solve descriptive problems.
Language data come from the author's judgements and the judgements of
other native speakers of the same variety about the
acceptability/possibility of the application of rules/processes in
certain environments. In addition, there was recourse to production
data collected both in experimental conditions and in daily
conversations or in the speech of the media. Restricting the analysis
to one variety -the EP variety spoken in the Lisbon area- ensures the
exhaustivity and consistency of the claims presented. Chapter 1 is an
introduction, including a brief description of the theoretical
backgrounds and a short review of the literature about the status of
the clitic group and of the Strict Layering Hypothesis in Prosodic
Phonology. Lexical Phonology is also mentioned, although the
literature review focuses mainly on Prosodic Phonology (Nespor &
Vogel, 1986, Selkirk, 1984, and Hayes, 1989). The chapter presents
the diagnostics for the prosodic word and extends to issues like
resyllabification and prosodic restructuring. Important for
understanding what follows are some assumptions the author
makes. (Throughout the remaining of the book, she is careful to supply
evidence that support these assumptions). First, it is assumed that
the Strict Layering Hypothesis may be relaxed, that is, if it is
decomposed in related statements (along the lines of Selkirk, 1995),
some of these statements may be violated. This assumption - which had
already been taken in Peperkamp (1997) - allows for recursive prosodic
structure that is argued for in chapters 5 and 6. Since an Optimality
Theory perspective is not assumed, these statements should be
understood as parameters, though this issue is not approached.
Second, relating to the lexical/postlexical distinction, the idea of
precompilation (Hayes, 1990) is adopted. This idea is applied to
explain ambiguous behavior of host + clitic sequences, as regards the
lexical/postlexical status. The other assumption is that (as has been
argued in Booij (1988) and others) prosodic structure up to the level
of the prosodic word is already present in the lexical component. This
has consequences for the analysis, since phenomena that refer to the
prosodic word may be either postlexical or lexical. Some diagnostics
for prosodic words based on lexical processes will be relevant only
for morphosyntactic structures derived in the lexicon, while other
diagnostics, based on postlexical processes, will be relevant only for
sequences obtained in the postlexical level. As regards the clitic
group, the author adopts the thesis that it is not a needed prosodic
category (Booij, 1988, 1996, Selkirk, 1995, and many others). Once the
clitic group is excluded from the prosodic hierarchy, several
alternative analyses of clitic plus host prosodic structure have
to be investigated. These analyses are presented in chapter 1 and discussed
carefully in chapter 5. Chapter 2 reviews previous studies on EP word
phonology. The author states that 'the prosodic word domain is almost
totally absent from the phonological descriptions of EP.' (p.62) So,
the disperse references to the word obtained from a miscellanea of
structuralist, generative and post- generative studies are presented
and discussed. One exception are the reflections of the Brazilian
linguist C�mara Jr. about the distinction between the
morphological and the phonological word, that Vig�rio thoroughly
reviews. Chapter 3 presents the phenomena that will be later on used
to diagnose prosodic constituency. Each phenomenon is briefly
described and then ascribed to the lexical/postlexical level of E
phonology. I reckon that it is important to learn about the phenomena
and to be presented to independent evidence for the lexical or
postlexical status of each of them. However, I do not believe that the
amount of phenomena and data discussed in this chapter will be of easy
reading to anyone not familiar with EP phonology. Some processes are
too different in nature to be treated in the same chapter, while
others are so similar that they can be easily confounded. Very helpful
is a chart that summarizes the division of rules into lexical and
postlexical levels at the end of the chapter. In Chapter 4 the issue
of the affix or clitic behavior of stressless pronouns is
approached. Stressless pronouns may occur inside inflectional
affixes in EP, as in 'perceber- TE-ia' 'note -YOU -Conditional/3rdperson
singular'. They may also trigger idiosyncratic segmental insertions
and deletions, as in 'come-lo' 'eat, present 2ndperson singular - it',
and, further, since they only admit verbs as their hosts, they show a
selectivity that is characteristic behavior of affixes. Contrary to
the position argued for in Zwicky (1987) and others that pronominal
clitics have been reanalyzed as inflectional affixes, Vig�rio
presents evidence that they are attached to the verb only in the
post-lexical level, while inflectional affixation is considered to
happen only in the lexicon. First, as far as word stress is concerned,
clitics do not modify their host's stress location. Second, there are
processes that do not apply to host + clitic sequences, while they
apply to base + suffix. One example: in EP /e/ changes to a half- open
central vowel when followed by a palatal segment; this centralization
applies across the morphological juncture in 'Europe + -izar' 'to turn
into European', but not in the host + clitic juncture in 'd�-lho'
'give it to him'. Since there is independent motivation for
considering centralization and other processes as lexical, they
constitute evidence that host + clitic sequences are obtained only
postlexically. Suffixes, on the other hand, are attached to their
base in the lexical level and are, therefore, input to stress
assignment and other lexical processes. The above-mentioned aspects
that point to a lexical status of these cliticizations -'mesoclisis',
segmental substitutions- are accounted for with precompiled rules and
allomorphy. Chapter 5 deals with the prosodic structure of
affixations and cliticizations in EP. Suffixes are shown to be
incorporated into their base's prosodic word, while monosyllabic
prefixes are adjoined to the word, yielding a recursive structure,
where one prosodic word is embedded in a higher-order prosodic word,
e.g. (RE(ORGANIZAR)w)w 'to reorganize' (similar to Peperkamp's
proposal for Spanish and Italian productive monosyllabic prefixes)
. This prosodic distinction results from a morphological distinction:
suffixes attach to stems while prefixes attach to words (or themes).
The same adjunction is argued for in rightward cliticization;
enclitics, on the other hand, are incorporated into their host's
prosodic word, like suffixes. Contrary to the idea put forward in
Brand�o de Carvalho (1989) that EP tends towards leftward
cliticization, Vig�rio takes phonological evidence to support
the view that only pronominal clitics are enclitic in EP, while other
functional words (definite articles, prepositions, complementizers)
behave consistently as proclitics. The evidence for proclitic status
of these functional words is the following: (i) the fact that they
behave in a way distinct from enclitics in several phonological
processes (for example, the definite article O does not semivocalize
when preceded by a vowel, as happens with the enclitic pronoun O
('him'), e.g. Eu vi o *[iw] texto ''I saw the text'' vs. Eu vi-o [iw]
deitado ''I saw him lying'' vs. (p. 186)); (ii) the fact that these
words, contrary to clitic pronouns, may appear in intonational phrase
initial position; (iii) the fact that these words may optionally
receive intonational phrase prominence and pitch accent. Although
enclitics are like suffixes incorporated into the preceding prosodic
word, similar to what Peperkamp (1997) has proposed for Lucanian and
Booij (1996) for Dutch, they present not exactly the same behavior as
suffixes as already mentioned. This at first sight should deny the
idea that EP enclitics bear the same prosodic structure as suffixes
do. However, as was stated above, Vig�rio contends that this can
be explained by the locus in the grammar where host + clitic sequences
are obtained: the postlexical level. Enclitics do not undergo exactly
the same processes as suffixes do, because the latter are added in the
lexical level. The same contrast is argued to hold between prefixes
and proclitics, although there are no lexical phonological phenomena
which may support this. The fact that also postlexically adjunction
happens at the left edge, while incorporation happens at the right, is
explained by the idea that only the left edge of lexical prosodic
words is visible when words are concatenated into the postlexical
level. Since clitics are not prosodic words by themselves, they are
incorporated into the prosodic word on the left (cliticization to the
left depends on particular syntactic configurations too) or else
adjoined to the right. Here, Vig�rio seems to adhere to a
end-based mapping approach (Selkirk, 1986) while, in the first
chapter, a relation-based mapping approach had been implicitly
adopted. In Chapter 6 the issue of the prosodization of compound
words is investigated. Vig�rio discusses if this prosodization
should be distinct from regular p-phrases. Taking into account,
besides regular morphological and syntactic compounds, a series of
words not generally considered as compounds, like derived words with
the diminutive suffix -ZINHO and the adverbialyzing suffix - MENTE,
abbreviations, letter names, mesoclitic constructions, etc.- she shows
that there is particular behavior that may point to a special prosodic
constituent for compounds. Evidences presented are the following:
segmental deletion rules, like non-back vowel deletion (e.g. doce[0]
�gua ''sweet water'') and optional back vowel deletion
(eg. salto[0] alto... ''high heels'') tend to be blocked in
compounds (e.g. onze avos ''eleventh'', porta- �culos ''glasses
holder''). Moreover, focal stress may only occur in the rightmost
prosodic word of compound words, while it may occur on any of the
prosodic words that constitute a p-phrase. Thus, a compound prosodic
word, a prosodic word that dominates two constituents of the same type
is argued for. I think that the idea for such a particular structure
should not be rejected a priori. However, it seems to me that there is
not sufficient empirical justification for it since the phenomena
adduced as evidence are too malleable. We are in need of more
research about phrasal processes that may highlight the issue.
Chapter 7 focuses on the characteristic reductions of clitics, which
are not generalizable to other unstressed syllables. Data collected in
experimental conditions are reported. The author concludes that the
idiosyncratic reductions of consonant-schwa clitics (preposition 'de',
complementizers and pronouns) and other non-schwa clitics (like the
prepositions EM in', COM 'with', PARA 'for', and others) should be
ascribed to the existence of lexically stored reduced
allomorphs. Non-reduced allomorphs, on the other side, may also be
affected by general reduction processes, which explains some gradient
effects that have been found in the analyzed data. Since this chapter
must be related to chapters 4 and 5, where clitics have been
discussed, it seems out of place here. Chapter 8 is an excellent
conclusion, summarizing the contents of the book as a whole,
underlining the results achieved, pointing to and discussing
alternative analyses, and raising questions of theoretical relevance
and empirical questions for future research. Overall, this book
constitutes a careful investigation of the prosodic phonology of the
level of the word in EP. It provides a lot of empirical data, and
some of the results are very interesting. The data are accounted for
in terms of distinct prosodizations within the domain of the prosodic
word allied with the lexical/postlexical distinction, which seems to
capture most of the data. As regards the proposal of a compound
prosodic word, I think it should have been motivated with more
substantive evidence, since, in practice, it adds a new kind of
constituent to the prosodic hierarchy. Except for some minor details
that were not completely clear to me, the issues are clearly presented
and thoroughly discussed. Therefore, this work will be of use for
everybody interested not only in the phonology of EP, but also in
phonological theory concerning Prosodic Phonology.


BOOIJ, G.(1988) On the relation between lexical and prosodic
phonology. In: BERTINETTO, P. M. And LOPORCARO, M. (eds.) Certamen
Phonologicum. Torino: Rosemberg and Sellier. p.63-76.

BOOIJ, G. (1996) Cliticization as prosodic integration: The case of
Dutch. The Linguistic Review, n.13.p. 219-242

BRANDfO de CARVALHO, J. (1989) Phonological conditions on
Portuguese clitic placement: on syntactic evidence for stress and
rhytmical patterns. Linguistics. n. 27. p. 405-436.

HAYES, B. (1989) The prosodic hierarchy in meter. In: KIPARSKY, P. and
YOUMANS, G. (eds.) Rhythm and Meter. Phonetics and Phonology I. New
York: Academic Press. p.201-260.

HAYES,B. (1990) Precompiled phrasal phonology. In: INKELAS, S. and
ZEC, D. (eds.) The Phonology-Syntax Connection. Chicago: Chicago
University Press. p. 85-108.

NESPOR, M. and VOGEL, I. (1986) Prosodic Phonology. Dordrecht,
Holland: Foris.

PEPERKAMP, S. (1997) Prosodic words. HIL Dissertations 34. The Hague:
Holland Academic Graphics.

SERKIRK, E. (1984) Phonology and Syntax. The Relation between Sound
and Structure. Cambridge, MA: MIT Press.

SELKIRK, E (1986) On derived domains in sentence phonology. Phonology
Yearbook, n. 3. p. 271-405.

SELKIRK, E.(1995)The prosodic structure of function
words.In: BECKMAN,J., DICKEY, L. W. and URBANCZYK, S. (eds.) Papers in
Optimality Theory. University of Massachusetts Occasional Papers,
n. 18. Amherst, MA: GLSA. p. 439-469.

ZWICKY, A. (1987) Suprressing the Z's. Journal of Linguistics. v.23,
n.1. p.133-148.


Gisela Collischonn is Adjunct Professor at Universidade Federal do Rio
Grande do Sul, Brazil, where she teaches phonology and morphology.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue