Romance Phonetics and Phonology

Mark Gibson
Visit to download the full and correct content document:
https://ebookmass.com/product/romance-phonetics-and-phonology-mark-gibson/
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

Romance Phonetics and Phonology

OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

Romance Phonetics
and Phonology

Edited by
MARK GIBSON AND JUANA GIL

1
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

3
Great Clarendon Street, Oxford, OX DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© editorial matter and organization Mark Gibson and Juana Gil 
© the chapters their several authors 
The moral rights of the authors have been asserted
First Edition published in 
Impression: 
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
 Madison Avenue, New York, NY , United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 
ISBN ––––
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

Contents
Acknowledgments vii
List of Figures and Tables ix
List of Abbreviations xvii
The Contributors xxi

. Romance sounds: New insights for old issues 

Mark Gibson and Juana Gil

Part I. Acoustic Studies

. Rhotic variation in Spanish codas: Acoustic analysis and effects of
context in spontaneous speech 
Beatriz Blecua and Jordi Cicres
. The phonetics of Italian anaphonesis: Between production
and perception 
Silvia Calamai
. A crosslinguistic study of voiceless fricative sibilants in Galician and
European Portuguese 
Xosé Luís Regueira Fernández and María José Ginzo
. Acoustic realization of vowels as a function of syllabic position:
A crosslinguistic study with data from French and Spanish 
Cédric Gendrot, Martine Adda-Decker, and Fabián Santiago

Part II. Articulatory Studies

. An articulatory account of rhotic variation in Tuscan Italian:
Synchronized UTI and EPG data 
Chiara Celata, Alessandro Vietti, and Lorenzo Spreafico
. Vowels and diphthongs: The articulatory and acoustic structure
of Romanian nuclei 
Ioana Chitoran and Stefania Marin
. Temporal organization of three-consonant onsets in Romanian 
Stefania Marin
. Articulatory setting, articulatory symmetry, and production
mechanisms for Catalan consonant sequences 
Daniel Recasens and Meritxell Mira
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

vi Contents

Part III. Studies in Perception

. Perceptual cues for individual voice quality 
Marianela Fernández Trinidad and José Manuel Rojo Abuin
. Perception of lexical stress in Spanish L by French speakers 
Joaquim Llisterri and Sandra Schwab
. Brazilian Portuguese rhotics in poem reciting: Perceptual, acoustic,
and meaning-related issues 
Sandra Madureira
. Perceived phrasing in French: A survey of some sentence structures 
Caroline L. Smith

Part IV. Phonological Issues

. Modeling assimilation: The case of sibilant voicing in Spanish 
Rebeka Campos-Astorkiza
. Adjusting to the syllable margins: Glides in Catalan and Spanish 
Jesús Jiménez, Maria-Rosa Lloret, and Clàudia Pons-Moll
. Galician mid-vowel reduction: A Stratal Optimality Theory account 
Fernando Martínez-Gil

Part V. Studies in Acquisition

. Language proximity and speech perception in young bilinguals:
Revisiting the trajectory of infants from Spanish–Catalan contexts 
Laura Bosch
. Production and perception in the acquisition of Spanish and Portuguese 
Jaydene Elvin, Polina Vasiliev, and Paola Escudero
. Production of French close rounded vowels by Spanish learners:
A corpus-based study 
Isabelle Racine and Sylvain Detey
. Phonetic behavior in proficient bilinguals: Insights from the
Catalan–Spanish contact situation 
Miquel Simonet

References 
Index 
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

Acknowledgments
Dozens of people were involved in the production of the current volume. Authors,
editors, and publishers all played a crucial role in the development and execution of
the final product. Add in the family members and support teams of all those
involved, and the number of dedicated people who have invested their time in this
book grows by the hundreds. However, of all the people involved who have made this
volume what it is, the editors are eternally indebted to the hard work, perseverance,
and dedication of one person, style and copy editor María Nuria Martínez García.
Nuria had the thankless job of dotting every ‘i’ and crossing every ‘t’, and performing
other odd tasks too numerous to mention here. We would like to formally thank her
for her outstanding and painstaking work. It was truly inspiring to work with
someone so meticulous and dedicated to her work.
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

List of Figures and Tables

Figures
. Spectrograms of the sequences terce(ro), por lo, and corer, showing
single-component rhotics (occlusion, approximant, and fricative) 
. Spectrograms of the sequences porD(ios) and tard(e), showing rhotics of
two components (approximant + vocalic element and occlusion + vocalic
element) 
. Spectrograms of the sequences porte(ría) and (for)mar la(zos), showing
three-component rhotics (approximant + vocalic element + approximant,
and occlusion + vocalic element + occlusion) 
. Spectrogram of the sequence contarlo, showing an example of elision
(indicated by Ø) 
. Spectrogram of the sequence (inten)tar eh. The rhotic has five components
(occlusion + vocalic element + occlusion + vocalic element + approximant) 
. Spectrograms of the sequences (cual)quier co(sa), joder, and hacer,
two-component rhotics with a fricative component (occlusion + fricative,
approximant + fricative, and fricative + vocalic element, respectively) 
. Spectrogram of the sequence cerc(a), with a three-component rhotic
(occlusion + fricative + approximant) 
. Percentages of the number of components according to position 
. Dispersion diagram of the analysis of the correspondence between the
number of components and place of articulation of the following consonant 
. Dispersion diagram of the analysis of the correspondence between the
number of components and manner of articulation of the following
consonant 
. Numbers of tokens of the different component variants according to
the manner of articulation of the following consonant 
. Percentages of realizations of the first component according to the
position and the number of components 
. Percentages of realizations of the third component according to position 
. Percentages of realizations of the second component in the realizations with
an occlusive first component according to their position 
. Box plots of the duration according to the number of components and
the position 
. Diagram of the average duration of rhotics (and of each of their components)
according to their number of components, position, and acoustic characteristics 
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

x List of Figures and Tables

. Average spectra for front sibilants (S) in female Galician (GF, grey) and female
Portuguese (PF, black) speakers 
. Average of the three spectrum types for male Galician (GM) speakers: S
(average of four speakers), S (average of two speakers), S (average
of two speakers), compared with male Portuguese [s] (PM-S) 
. Average spectra of the front sibilant of female Galician (GF) informants
from the apical [s̺] area and the laminal [s] area 
. Average final sibilant (PSF) and syllable-initial [ʃ] (PX) spectra in realizations
by Portuguese speakers, female (F) and male (M) groups 
. Average spectra of [s̪] (GM–G–S), [ʃ] (GM–G–X), and the sibilant in
coda position (GM–G–SF), in the two Galician speakers with
lamino-dental seseo 
. Average spectra for realizations of [s] (GF–S), [ʃ] (GF–X), and word-final
sibilant (GF–SF) in female Galician speakers (GF) 
. Normalized vowel formants (F and F) in normalized Hz for Spanish
(left) and French (right) 
. Normalized acoustic dispersion (in normalized Hz) for Spanish (left)
and French (right) 
. Normalized acoustic dispersion (in normalized Hz), duration
(in normalized ms), and f₀ (in normalized Hz) values (from left to right) of vowels
in initial and final syllables in French disyllabic words 
. Normalized acoustic dispersion (in normalized Hz), duration
(in normalized ms), and f₀ (in normalized Hz) values (from left to right) of
vowels in initial unstressed syllables compared to final stressed syllables
in Spanish disyllabic words 
. Normalized acoustic dispersion (in normalized Hz), duration (in normalized ms),
and f₀ (in normalized Hz) values (from left to right) of vowels in initial stressed
syllables compared to final unstressed syllables in Spanish disyllabic words 
. Normalized formants of vowels (in normalized Hz) for French and Spanish
according to the presence of pause (the lower line for final prepausal
vowels, the upper line for final nonprepausal vowels) 
. Distribution of the tense–lax feature for three alveolar sonorants in central
Europe and the Italo-Romance domain 
. Percentages of rhotic variants in the corpus (by subject) 
. Wiring diagram 
. Results of the MCA of the rhotic variants: plane representation
of the category cloud 
. Constriction duration values, F and F values at consonant onset and
offset for the singleton vs. geminate contrast as a function of vowel context 
. Constriction Qp (left), CAa (middle), and CCa (right) for the singleton vs.
geminate contrast as a function of vowel context 
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

List of Figures and Tables xi

. Smoothing spline estimates and % Bayesian confidence interval for
comparison of the mean curves for /r/ in /ar(ː)ra/, /ur(ː)u/, and /ir(ː)i/ for
subjects CHB (a) and CHC (b) 
. Smoothing spline estimates and % Bayesian confidence interval for
comparison of the mean curves for /r/ in /ara/ vs. /arːa/ (a), /iri/ vs. /irːi/ (b),
and /uru/ vs. /urːu/ (c) for subject CHB (singletons: VRV; geminates: VRRV) 
. Smoothing spline estimates and % Bayesian confidence interval for
comparison of the mean curves for /r/ in /ara/ vs. /arːa/ (a), /iri/ vs. /irːi/ (b),
and /uru/ vs. /urːu/ (c) for subject CHC (singletons: VRV; geminates: VRRV) 
. Dynamic movement of the singleton /r/ (a) and the geminate
(b) in the /i/ context for speaker CHB 
. Dynamic movement of the singleton /r/ (a) and the geminate (b) in the /i/
context for speaker CHC 
. Example measurement of vowel articulatory movement for vowel /e/ in
diphthong /ea/ in one repetition of the word /ka.ˈfea/ 
. Vowel-to-vowel timing lag, representing the lag between maximal constriction
of the two vowels 
. Position of the first vowel/glide (V) in diphthong/hiatus sequences
and position of vowel /e/ 
. Position of the second vowel (V) in diphthong/hiatus sequences and position
of vowel /a/ 
. Difference in position between the two vowel targets (V–V) at the point
of maximum constriction for each 
. Top: Diphthong /ja/ in [ˈpja.trʌ]. Bottom: Hiatus /i.a/ in [pi.ˈa.stru] 
. Mean and one standard deviation for acoustic duration (left) and F values
at vowel-onset time point (right) as a function of category 
. Scatterplot of mean durations by mean F as a function of speaker 
. Schematic representation of the timing changes predicted by a c‑center
organization as a function of onset complexity increase 
. Average lags of the vowel-adjacent consonant to the anchor 
. Average lags of the vowel-adjacent consonant to the anchor as a function of set 
. Averages of absolute (top) and normalized (bottom) intra-cluster timing
for three consonant clusters, between consonants  and  (CC) and
between consonants  and  (CC) 
. Mean COG values and COG ranges for fricatives and fricative + fricative
sequences (left) and for affricates and affricate + fricative sequences (right)
in Eastern, Western, and Valencian Catalan 
. Linguo-palatal contact patterns for fricatives and fricative + fricative
sequences taken at the midpoint of the frication noise according to the
Eastern Catalan speakers DR, JP, and JC 
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

xii List of Figures and Tables

. Linguo-palatal contact patterns for affricates and affricate + fricative

sequences taken at the midpoint of the closure and frication phases according
to the Eastern Catalan speakers DR, JP, and JC 
. Discrimination rate by speaker (indicated by ‘S’) calculated
using the quotient of hit responses 
. Intra- and interspeaker variation by parameter group (indicated by a letter) 
. Inter- and intraspeaker variation by speaker and parameter group
(indicated by a letter) 
. Changes in parameters per speaker when shifting registers (modal–falsetto) 
. Average utterance length per speaker (indicated by ‘S’ and its corresponding
number) and register 
. Spectrogram of a sample (Dica dadiva) in modal voice by Speaker  
. Spectrogram of a sample (Dica dadiva) in modal voice by Speaker  
. Percentages of correct identification rates for Base stimuli as a function of
the stress pattern and the competence of French participants in L
Spanish (Advanced, With no knowledge) 
. Percentages of correct identification rate by French listeners for Manipulated
stimuli as a function of the seven manipulations and the competence in L
Spanish (Advanced, With no knowledge) 
. Percentages of different (Diff) responses by French listeners as a function
of the pair member (PP > P paired with PP, PP > P paired with P;
PP = proparoxytone; P = paroxytone) and the competence in L Spanish
(Advanced, No knowledge) for the seven possible manipulations 
. Percentages of different (Diff) responses by French listeners as a function
of the pair member (P > O paired with P, P > O paired with O; P = paroxytone;
O = oxytone) and the competence in L Spanish (Advanced, No knowledge)
for the seven possible manipulations 
. Percentages of correct responses in the training session as a function
of the groups (Natives and Non-natives) and of the five training blocks 
. Estimated probability of correct responses in the post-test as a function of the
percentages of correct responses in the training session (in Non-natives only) 
. A voiceless glottal fricative produced in syllable coda position of the
word mar (‘sea’) 
. An approximant produced in syllable coda position of the word mar (‘sea’) 
. A voiced alveolar tap produced at the syllable coda position of the word
mar (‘sea’) 
. A voiceless fricative trill produced at the syllable coda position
of the word mar (‘sea’) 
. A voiced trill produced at the syllable onset position of the word
arranca (‘rip’) 
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

List of Figures and Tables xiii

. Dendrogram showing the grouping of the stanzas in clusters 

. The distribution of the groups of variables Gq (VPAS), Gc
(ExpressionEvaluator), and the perceptual evaluation of activation
(ZAP, Gc) in two dimensions (Dim  and Dim ) of the vector space 
. The projection onto two dimensions of the quantitative factor variables 
. Comparison of b-scores and p-scores for read and spontaneous dislocations 
. A sentence (‘She read a novel in Arabic.’) produced in broad-focus
condition by a female speaker, showing the waveform (above) and f₀ trace 
. A sentence produced in narrow-focus condition by a male speaker, showing
the waveform (above) and f₀ trace 
. Comparison of scores for words in different positions before and during the
focus phrases, under broad and narrow focus 
. Scores for words ending focus phrases, comparing those that are
sentence-medial to those that are sentence-final 
. Comparison of the scores for the final word in long- or short-subject noun
phrases (NPs) 
. Spectrogram and waveform illustrating the three acoustic measurements,
i.e., vowel duration, voicing during the fricative, and fricative duration for
the word /atisˈbe/ 
. Distribution of the percent voicing for tokens with voiceless sequences,
i.e., /s/ followed by a voiceless consonant (Experiment ) 
. Distribution of the voicing categories of /s/ before a voiced obstruent for each
speaker 
. Close (top panel) and open (bottom panel) approximant realizations
of the voiced obstruent in the word /atisˈbe/ 
. Histogram for the distribution of percent voicing in the intonational
phrase boundary condition 
. Distribution of the percent voicing for tokens with voiceless sequences
(Experiment ) 
. Histogram for the distribution of percent voicing before a voiced obstruent
in the voicing condition in Experiment  (left) and Experiment  (right) 
. Distribution of the voicing categories by speaker for the voicing condition
in Experiment  (left) and Experiment  (right) 
. Realization of /w/ in huelo and cacahuete in Castilian Spanish 
. Realization of /j/ in yugo, (el) yerno, and mayo in Castilian Spanish 
. Realization of /w/ in diuen in Majorcan Catalan 
. Realization of /j/ in deia in Majorcan 
. Praat screenshot of a sound file and a TextGrid file 
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

xiv List of Figures and Tables

Tables
. Number of rhotic consonants before a consonant and before a pause
per speaker 
. Acoustic characteristics used in the definition of the components 
. Factors and variants 
. Composition of the rhotic consonants according to the characteristics
of each of their components, and the number of realizations found
in the corpus 
. Percentage of realizations in spontaneous and read speech, according to
number of components 
. Scale of acceptability of velar clusters 
. Percentage of the answers according to the factors Place and Vowel before
the velar cluster 
. Percentage of the answers according to the factors Place, Vowel, and
Postnasal consonant 
. Confusion matrix for the alveolar clusters–high vowels (control
items–percentage values) 
. Confusion matrix for the velar clusters–high vowels (control
items–percentage values) 
. Confusion matrix for the alveolar clusters–mid-high vowels
(test items–percentage values) 
. Confusion matrix for the velar clusters–mid-high vowels (test
items–percentage values) 
. Development of the sibilant system from medieval Galician-Portuguese
to modern standard Portuguese and Galician 
. Spectral mean and standard deviation (all informants) for each sibilant
(in onset position) separated by sex 
. Spectral means, standard deviations for alveolar and postalveolar
fricatives in Galician and Portuguese male and female groups 
. Spectral mean, kurtosis, and skewness for Galician male (GM) groups
S, S, and S 
. Correlations between fricatives for spectral mean in Galician speakers by sex 
. Correlations between fricatives for spectral mean in Portuguese speakers
by sex 
. Frequency of rhotic variants in the corpus (by subject) 
. List of variables and corresponding values associated with the /r/
tokens in the MCA 
. Average values (in Hz) and standard deviation of F and F at the acoustic
onset and offset of the constriction phase (single-constriction rhotics) 
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

List of Figures and Tables xv

. Effects of length (singleton vs. geminate), vowel context (/i/ vs. /a/ vs. /u/),
and length by vowel context on constriction duration, and F and F
constriction onset and offset values 
. Effects of length (singleton vs. geminate), vowel context (/i/ vs. /a/ vs. /u/),
and length by vowel context on constriction Qp (left), CAa (mid),
and CCa (right) 
. Target stimuli for the articulatory study, with intervals of interest
shown in boldface 
. Target stimuli for the acoustic study, with intervals of interest shown in
boldface 
. Stimuli 
. Sentence list with the consonants and consonant sequences under analysis
underlined, and the syllable bearing sentence stress in boldface 
. Mean, highest, and lowest COG values, and COG ranges, plotted in
Figure . 
. Mean, standard deviation, standard error of the mean, and confidence
intervals for d-prime (fifty-seven listeners) 
. Number of glottal parameters (in groups) that each speaker varied
significantly (p < .*) when shifting registers (modal–falsetto) 
. Poem parts, their themes, and percentage of rhotic varieties produced in
syllable onset position 
. Poem parts, their themes, and percentage of rhotic varieties produced
in syllable coda position 
. Factor variables and their correlation coefficients and p-values 
. Positions of words identified for analysis, with number of words analyzed
in each position and abbreviation codes 
. Means (ms) for preceding vowel duration, fricative duration, and percent
voicing, with standard deviations in parentheses 
. Distribution of voicing categories depending on the voicing of the following
obstruent 
. Distribution of following consonant manner as a function of stress 
. Mean and standard deviation (in parentheses) for the percent voicing as
a function of stress and manner of the following consonant 
. Distribution of voicing categories by following approximant type 
. Mean vowel and fricative duration (ms) and percent voicing
for each boundary type 
. Distribution of voicing categories by prosodic boundary type 
. Sample stimuli for the stress-pattern condition (c) 
. Distribution of voicing categories of /s/ following consonant manner 
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

xvi List of Figures and Tables

. Mean vowel duration (ms) according to stress pattern 

. Mean fricative duration (ms) by stress pattern and following
consonant manner 
. Mean percent voicing by stress pattern and following consonant manner 
. Distribution of voicing categories according to stress pattern with a
following obstruent (top) and a following sonorant (bottom) 
. SLM, PAM and PAM-L, and LLP models’ focus and predictions
for L success 
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

List of Abbreviations
 AFC two-alternative forced-choice experiment
A approximant
AAA Articulate Assistant Advanced software
AE American English
ALE Atlas Linguarum Europae
ALeCMan Atlas lingüístico etnográfico de Castilla-La Mancha
ALiR Atlas Linguistique Roman
ALPI Atlas Lingüístico de la Península Ibérica
ALPS Alpine Laboratory of Phonetic Sciences
ANOVA Analysis of Variance
AP accentual phrase
ATR Advanced Tongue Root
AusE Australian English
AV audiovisual
BFLA bilingual first language acquisition
BP Brazilian Portuguese
C consonant
C second consonant
CAH Contrastive Analysis Hypothesis
CCA contact centrality in the anterior palate
CE Californian English
CEFRL Common European Framework of Reference for Languages
CETENFolha Corpus de Extractos de Textos Eletrônicos of the Folha de São Paulo newspaper
CG Category Goodness
CGal Conservative Galician
COG center-of-gravity
CSL Computerized Speech Lab (data acquisition system)
ED Euclidean Distances
EMA electromagnetic articulography
EPG electropalatography
ERP event-related brain potential
ETRURiaS Electropalatographic and Ultrasound tongue imaging Rhotics Synchronized
corpus
F fricative
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

xviii List of Abbreviations

f₀ fundamental frequency
FFT fast Fourier transform
FOLERPA Ferramenta Online de Experimentación Perceptiva
GF Galician female
GM Galician male
HSD honest significant difference
Hz Hertz
IGal Innovative Galician
IP Intonational Phrase
ip intermediate phrase
IS Iberian Spanish
IPFC InterPhonologie du Français Contemporain
L first language
L second language
LLP Second Language Linguistic Perception
LD left dislocation/left dislocated
LIMSI Laboratoire d’Informatique pour la Mécanique et les Sciences de l’Ingénieur
LMEDS Language Markup and Experimental Design Software
LPC linear predictive coding
LSC Lax-to-Stress Condition
LT laryngeal tenseness
MCA multiple correspondence analysis; multiple-category assimilation
MFA multiple factor analysis
MRI magnetic resonance imaging
MVR mid-vowel reduction
N nasal
NCCF Nijmegen Corpus of Casual French
NP noun phrase
O oxytone
OCP obligatory contour principle
OO output-to-output
OT Optimality Theory
P paroxytone
PAM Perceptual Assimilation Model
PCA Principal Component Analysis
PF Portuguese female
PFC Phonologie du Français Contemporain
PM Portuguese male
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

List of Abbreviations xix

PP proparoxytone
PRIMIR Processing Rich Information from Multidimensional Interactive
Representations
PW prosodic word
PX Portuguese fricative (a form used simply for coding in this research)
QMU Queen Margaret’s University
R rhotic
R row of electrodes
RD right dislocation/right dislocated
RPT Rapid Prosody Transcription
SCA Single-Category Assimilation
SF sibilants produced by females (a form used simply for coding in this research)
SLM Speech Learning Model
SNS Scuola Normale Superiore (Pisa, Italy)
T tap
TB tongue body
TCA Two-Category Assimilation
Tr trill
TV theme vowel
TVT vocal tract tenseness
UTI ultrasound tongue imaging
V vowel
VL variation of loudness
VOT voice onset time
VPAS Voice Profile Analysis Scheme
VRRV vowel-rhotic-rhotic-vowel
VRV vowel-rhotic-vowel
VV vowel to vowel
XAB refers to a type of perception test with two options (A, B) for stimuli
ZAP perceptual evaluation of activation
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

The Contributors
M ARTINE A DDA -D ECKER holds an MD in Applied Mathematics and a PhD in Computer
Science from the University Paris-Sud (Orsay), France. She has been a CNRS researcher since
. She joined the Laboratory of Phonetics and Phonology (LPP, UMR ) in , after
her previous position dealing with the Spoken Language Processing group at LIMSI-CNRS
(Orsay), where she remains an associate researcher. Her research interests focus on man-
machine communication, language and accent identification, multilingual speech recognition,
acoustic-phonetic and lexical modeling, pronunciation variants, phonetics, phonology, and
large corpus-based studies. She has authored or co-authored over  peer-reviewed articles in
the field. She is regularly reviewing papers in the major speech-related journals and confer-
ences. She collaborates as an expert in French national, Belgian, Canadian, Luxembourgish,
Swiss, and European research funding agencies.

B EATRIZ B LECUA is a Professor in Spanish Phonetics and Phonology at the University of

Girona. She obtained her PhD in Linguistics in  from the Autonomous University
of Barcelona (UAB), doing her dissertation on rhotic sounds in Spanish, which continues to
be one of her main research interests. Her areas of specialization include Spanish phonetics,
acoustic phonetics, phonetic variation in speaking styles, and its relation with sound change,
and pronunciation problems in second-language acquisition. Her current research focuses on
the perceptual effects of phonetic variation, as well as on interspeaker variation related to
forensic phonetics.
L AURA B OSCH is Professor in Psychology at the University of Barcelona (UB) and member of
the Institute for Research on Brain, Cognition and Behavior (IRC) of this institution. Her
research interests are focused on early speech perception, language learning, and phonological
development, in both monolingual and bilingual populations. She coordinates the infancy
research lab at the University of Barcelona (APAL, Attention, Perception and Acquisition of
Language) and her current research also includes populations at risk of neurocognitive and
language-based disorders.

S ILVIA C ALAMAI received her PhD in Linguistics from the University of Perugia in . She is
now Associate Professor in Linguistics at the University of Siena. Her areas of research include
experimental phonetics, sociolinguistics, dialectology, and Intangible Cultural Heritage. At
present, she is the scientific co-coordinator of the Project Grammo-foni. Le soffitte della voce
(Gra.fo) at the University of Siena (http://grafo.sns.it), and she is Associate Member of the
Intellectual Property Issues in Cultural Heritage (IPinCH) International Project (http://www.
sfu.ca/ipinch/). List of publications at https://sites.google.com/a/unisi.it/silviacalamai/home.
R EBEKA C AMPOS -A STORKIZA works on phonetics and phonology, from both a theoretical and
an experimental perspective. Her work illustrates how phonetic data can help us develop
theoretical models to explain sound patterns. In addition, she is interested in second-language
acquisition of phonology, with a focus on new methodologies and instructional implications.
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

xxii The Contributors

Her mentoring experience has also allowed her to develop expertise in sociophonetics,
especially on how phonetic analysis and understanding can shed light on phonetic variation
and vice versa.
C HIARA C ELATA is a researcher in Linguistics at the Scuola Normale Superiore (SNS) in Pisa,
Italy, where she teaches graduate-level courses in experimental phonetics and phonology. She
also coordinates the research activities in Speech Production and Perception at the SNS
Linguistics Laboratory. She is currently running two research projects, one looking at rhotic
variation in selected Italian varieties as a function of phonological and socio-communicative
variables, and the other, funded by the ESF and French ANR in collaboration with other
European universities, seeking to develop a psycho-computational account of the phonotactic–
morphology interface in several Romance and Germanic languages.
I OANA C HITORAN is Professor of Linguistics at the Université Paris Diderot, Clillac-ARP, and
CNRS—Laboratoire de Phonétique et Phonologie (UMR ). Her research focuses on the
phonetics–phonology interface, and the relation between temporal variability and phono-
logical structure, particularly in Caucasian and Romance languages.
J ORDI C ICRES is a Lecturer in language didactics and forensic phonetics at the University of
Girona. He holds an MA in Forensic Linguistics and a PhD in Applied Linguistics (University
Pompeu Fabra). His main research interests deal with descriptive phonetics both at the
segmental and at the suprasegmental levels, forensic linguistics (especially with applications
to speaker identification, transcription of disputed recordings, and the construction of linguis-
tic profiles), language variation, and general language didactics. His recent work has focused on
the analysis of Spanish rhotics and filled pauses with both descriptive and applied purposes
related to speaker identification.

S YLVAIN D ETEY is Associate Professor of Applied Linguistics and French Studies at Waseda
University (Japan), formerly Maître de Conférences (Assistant Professor) in the Department of
Language Sciences and Communication at the University of Rouen (France). His current
research interests lie in the use of oral corpora for language education and the role of variation
and multimodality in second-language phonology acquisition. He is one of the coordinators of
the (Inter)Phonologie du Français Contemporain (IPFC) project, and co-edited Les variétés du
français parlé dans l’espace francophone. Ressources pour l’enseignement (, Ophrys) and
Varieties of Spoken French (, Oxford University Press).
J AYDENE E LVIN received her PhD from The MARCS Institute, Western Sydney University. Her
research focuses on second-language speech perception, word recognition and production,
with a particular focus on Australian English and Iberian Spanish learners of Brazilian
Portuguese.
P AOLA E SCUDERO is Associate Professor at the MARCS Institute, Western Sydney University.
Her research focuses on speech and visual development in diverse populations, including
human infants, children and adults, and zebra finches.
M ARIANELA F ERNÁNDEZ T RINIDAD holds a Master’s degree in Phonetics and Phonology from
the Spanish National Research Council (CSIC)–the Menéndez Pelayo International University
in Madrid, Spain. She is currently a researcher at the CSIC Phonetics Laboratory, where she is
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

The Contributors xxiii

developing her doctoral thesis and performs research, both focusing on the perception and
production of voice in the fields of Phonic Science and Forensic Phonetics. She has been and
continues to be a member of various research teams financed by the Spanish government. In
Uruguay, she has participated in projects financed by the Sectorial Commission for Scientific
Research (CSIC) and has been a researcher with the National Research and Innovation Agency
(ANII). She is author of several articles published in both Spanish and international journals.

C ÉDRIC G ENDROT is Maitre de Conferences (Assistant Professor) at the University Sorbonne

Nouvelle—Paris . His early work focused on articulatory prosody: the synchronous physio-
logical and acoustic analysis of phonemes according to their prosodic position. Recently, he has
specialized in large corpora analyses of several languages, trying to relate previous results from
the physiological data from small corpora with large acoustic data in order to evaluate
tendencies and individual strategies in speech. Over the past two years, he has coordinated a
number of research projects—including an ANR JCJC project involving a comparison between
physiological and acoustic data.
M ARK G IBSON is Professor of Linguistics and Director of the Speech Laboratory at the
University of Navarra. His research focuses on articulatory timing in syllables from a Labora-
tory Phonology approach. His work mainly focuses on the timing of articulatory gestures in
syllables using electromagnetic articulography, ultrasound imaging, and aerodynamics.
J UANA G IL is currently Director of the Instituto Cervantes in Lyon, having previously held
positions at the Universidad Autónoma de Madrid and at UNED, the Spanish open university.
She also founded the Postgraduate Program in Speech Sciences organized by the Spanish
National Research Council (CSIC). She is mainly interested in the phonetics–phonology
interface and in some applications of phonetics, such as forensic phonetics and second-
language pronunciation learning and teaching. She is the co-editor, with Ricardo Mairal, of
Linguistic Universals (Cambridge University Press, ).
M ARÍA J OSÉ G INZO is Researcher and Statistical Consultant in Statistical Consulting Services
(SCS) at the University of Santiago de Compostela (USC). Her research is concerned mainly
with statistical techniques in Geolinguistics and Onomastics Modeling. She is co-author of
several papers with Spanish researchers on different topics connected with her work as
Statistical Consultant and this experience is reflected in many contracts with government
bodies and/or private companies. She has been a member of organizing committees for several
international conferences and scientific gatherings.
J ESÚS J IMÉNEZ is Associate Professor in the Department of Catalan Philology at the Universitat
de València, where he completed his PhD in  with a study on the syllable structure of
Catalan. He is also a member of the research center Institut Interuniversitari de Filologia
Valenciana. His current research focuses on phonology, phonetics, and linguistic variation in
Catalan, Spanish, and other Romance languages. He is the author of the book L’estructura
sil·labica del català () and a contributing author in the Gramàtica del català contemporani,
 vols. (, co-directed by Joan Solà, Maria-Rosa Lloret, Joan Mascaró, and Manuel Pérez
Saldanya). His work has appeared in journals such as Caplletra, Estudios de Fonética Experi-
mental, Estudis Romànics, Revista de Filología Románica, Rivista di Linguistica, and Verba.
Since , he is a member of a research group on the study of dialect variation led by
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

xxiv The Contributors

Maria-Rosa Lloret, from the Universitat de Barcelona (Grup d’Estudi de la Variació Dialectal
(GEVaD), ww.ub.edu/GEVAD).
J OAQUIM L LISTERRI received a PhD in Romance Philology from the Universitat Autònoma de
Barcelona (UAB) in , and he was appointed full-time lecturer in General Linguistics at
the same university in . He teaches courses in General Linguistics, General and Applied
Phonetics, Applied Linguistics, and Speech Technologies. His research, publications, and
participation in national and international projects have been concerned with the applica-
tion of experimental phonetics to speech technology—especially in the area of text-to-speech
synthesis—and with the study of phonetic transfer in second- and third-language acquisi-
tion. He has worked on the evaluation of synthesized speech and on the creation and
annotation of spoken databases and corpora, particularly in the domain of prosody. He
has also been involved in the use of communication and information technologies for
language teaching and for linguistic research.
MARIA-ROSA LLORET is Professor in the Department of Catalan Philology at the Universitat de
Barcelona. She completed her PhD in Linguistics at Indiana University in , with a study
on the morphophonology of Oromo (Cushitic). Her current research focuses on phonology,
morphology, and linguistic variation in Catalan, Spanish, and other Romance languages. She
is the author of the book La fonologia de català () and co-author of Manual de
transcripció fonètica (, with Eulàlia Bonet and Joan Mascaró) and Fonologia catalana
(, with Eulàlia Bonet). She co-directed the Gramàtica del català contemporani,  vols.
(, with Joan Solà, Joan Mascaró, and Manuel Pérez Saldanya). Her work has appeared in
journals such as Language, Lingua, Linguistic Inquiry, Phonology, Probus, Dialectologia et
Geolinguistica, Revista de Filología Románica, Catalan Journal of Linguistics, Caplletra,
Estudios de Fonética Experimental, and Verba. Since , she has led a research group on
the study of dialect variation (Grup d’Estudi de la Variació Dialectal (GEVaD), ww.ub.edu/
GEVAD).
S ANDRA M ADUREIRA has a PhD in Applied Linguistics from Pontifícia Universidade Católica
de São Paulo (PUC-SP), and performed post-doctoral work at the LAFAPE-IEL- (Laboratory
of Phonetics and Psycholinguistics–Institute of Language Studies) UNICAMP. She is Full
Professor in the Department of Linguistics at PUC-SP, lecturer at PEPG (Graduate Studies
Program) in Language Science Studies, and a researcher at the LIAAC (Integrated Laboratory
of Acoustic Analyses). She is head of the Research Group on Speech Studies and the editor of
the Journal Intercâmbio at PUC-SP and of the Journal of Speech Sciences maintained by LBASS
(Luso-Brazilian Association of Speech Sciences). Her main line of research is in Experimental
Phonetics. Her main areas of interest are: Speech Expressivity, Voice Quality, Acquisition of L
Sounds and Prosody.
S TEFANIA M ARIN was born and raised in Romania, where she obtained a BA (English and
Portuguese Philology) and MA (Applied Linguistics) from the University of Bucharest. She
received her PhD in Linguistics in  from Yale University, USA, and since then she has
been working at the Institute of Phonetics and Speech Processing at Ludwig Maximilian
University in Munich. Her research investigates how linguistically relevant structures such
as syllables are realized temporally at the speech-production level. The particular focus of her
work is on Romanian vowels and consonant clusters.
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

The Contributors xxv

FERNANDO MARTÍNEZ-GIL received his BA in Anglo-Germanic Philology from the University of

Salamanca, Spain, and his MA and PhD from the University of Southern California. He is an
Associate Professor of Spanish Linguistics in the Department of Spanish and Portuguese at Ohio
State University in Columbus, Ohio. He taught previously at Georgetown University. He has
(co-)edited several volumes on Spanish and Hispano-Romance phonology, including Issues in
the Phonology and Morphology of the Major Iberian Languages (, with Alfonso Morales-Front)
and Optimality-Theoretical Studies on Spanish Phonology (, with Sonia Colina). His main
research interests are Spanish and Galician phonology, phonological theory, and the historical
evolution of Spanish and the Hispano-Romance languages.
MERITXELL MIRA studied Chemical Engineering at the Universitat Politècnica de Catalunya. She
began her professional career as an engineer working in the field of big data processing before
moving to the Universitat Autònoma de Bellaterra as a research technician. There, she
collaborated with Dr. Daniel Recasens in the phonetics laboratory of the Institut d’Estudis
Catalans, performing experimental studies in phonetics and sound change. She currently
works as an engineer in the pharmaceutical industry.
C LÀUDIA P ONS -M OLL is a Tenured Professor in the Department of Catalan Philology at the
Universitat de Barcelona, where she completed her PhD in Catalan Philology in  with a
descriptive and formal study on Balearic Catalan phonotactics. Her current research focuses on
the phonology, the morphophonology, and the linguistic variation of Catalan and other
Romance languages from a theoretical perspective. Her work has appeared in journals and
edited series such as Acta Linguistica Hungarica, Linguistic Inquiry, Phonology, Probus,
Proceedings of NELS, Studies in Language, Current Issues in Linguistic Theory, Catalan Journal
of Linguistics, Caplletra, etc. She is the author of the book La teoria de l’optimitat. Una
introducció aplicada al català de les Illes Balears (), and she is co-directing, along with
Josefina Carrera, the project “Els sons del català” (http://www.ub.edu/sonscatala/). She is a
member of the Grup d’Estudi de la Variació Dialectal (GEVaD) (UB, PI: Maria-Rosa Lloret)
and of the Center for Theoretical Linguistics (UAB).

I SABELLE R ACINE is Professor of French as a Foreign Language at the University of Geneva,

Switzerland. Her research focuses on L phonological acquisition and on the implications of
phonetic and phonological variation at a pedagogical level. She has also published several
papers, book chapters, and conference papers on Swiss French. She is one of the coordinators
of the (Inter)Phonologie du Français Contemporain (IPFC) project, and co-editor of L’appren-
tissage de la liaison en français par des locuteurs non natifs: éclairage des corpus oraux (,
VALS-ASLA).
D ANIEL R ECASENS is Full Professor of Catalan Philology at the Universitat Autònoma de
Barcelona and director of the Phonetics Laboratory of the Institut d’Estudis Catalans. He
holds a PhD in Linguistics from the University of Connecticut and has carried out doctoral and
postdoctoral research at Haskins Laboratories (New Haven, CT, USA). He has published
numerous scientific articles in highly ranked international journals, several books and book
chapters, and is currently Associate Editor of Phonetica and member of the Editorial Board of
the Journal of the International Phonetic Association. He has served as Chair of the th
International Congress of the Phonetic Sciences () and as Vice-President of the Inter-
national Phonetic Association (–). His research interests include the interarticulatory
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

xxvi The Contributors

mechanisms in speech production, the phonetic causes of sound change, and the phonetics–
phonology interface.
XOSÉ LUÍS REGUEIRA FERNÁNDEZ is Professor of Galician and Portuguese Language and Literature
at the University of Santiago de Compostela. His research interests focus mainly on Galician
phonetics and sociolinguistics (standardization, variation, and change). He has authored
Dicionario de pronuncia da lingua galega [Dictionary of Galician Pronunciation] (A Coruña,
), co-authored Gramática galega [Galician Grammar] (Vigo, ), directed and co-
authored Os sons da lingua [The Sounds of Language] (Vigo, ], among other books. He
has published numerous articles on these topics in journals such as JIPA, Revista Internacional
de Lingüística Iberoamericana, Estudos de Lingüística Galega, Estudios de Fonética Experimen-
tal, Verba, Revista de Filología Románica, as well as many book chapters. He is a research fellow
at the Instituto da Lingua Galega [Galician Language Institute] and a member of the Real
Academia Galega [Royal Galician Academy].
J OSÉ M ANUEL R OJO A BUIN holds a bachelor’s degree in Statistics and a Master’s degree
in Science and Technology Statistics from the Universidad Complutense of Madrid. He is
the head of the Statistical Analysis Unit of the Center for Human and Social Sciences at the
Spanish National Research Council (UAE, CCHS, CSIC), Madrid, Spain. He participates in
the statistical design and analysis of research projects that are executed within the Institutes
and Research Groups in the CCHS. In this context, he is co-author of several books, chapters,
and papers published in national and international journals. He is a member of the Spanish
Society of Statistics and Operations Research (SEIO).

F ABIÁN S ANTIAGO completed a PhD in Descriptive, Theoretic, and Automatic Linguistics at

the University of Paris Diderot (Sorbonne Paris Cité) in . His research and publications
are mainly concerned with sentence phonology in both French and Spanish and the acquisition
of prosody (intonational and metrical aspects) of French as a foreign language. Recently, he has
worked on the applications of syntax–prosody mapping to automatic speech recognition
errors, hybrid phonetic annotations of multilingual synthetic voices, and the use of physio-
logical data (ultrasound) for modeling the articulation of the phoneme /r/ in French. He is
currently a postdoctoral research fellow at the Laboratory of Phonetics and Phonology at the
Université Sorbonne Nouvelle.
S ANDRA S CHWAB is a Lecturer in French phonetics at the University of Geneva. At the same
time, she is working on a SNF Ambizione project at the Phonetics Laboratory at the University
of Zurich. Her research focuses mainly on prosody in L and L. More specifically, she has
published experimental studies on speech rate in L and L, on regional prosodic variation in
French, and on the perception of Spanish lexical stress by French speakers.

M IQUEL S IMONET is Associate Professor of Spanish Linguistics at the University of Arizona. At

Arizona, he is a member of the faculty in the Department of Spanish and Portuguese, and he is
affiliated with the Department of Linguistics, the Program in Cognitive Science, and the
Program in Second Language Acquisition. He is an experimental phonologist and phonetician
whose main goal is to comprehend the effects of language use on the mental representations of
sounds, the shape of sound patterns, and the structure of sound systems. Much of his research
has been focused on bilingualism, second language speech, and sound change.
 OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi

The Contributors xxvii

C AROLINE L. S MITH is Associate Professor of Linguistics at the University of New Mexico.

She received her PhD from Yale University; while a student she worked as a research assistant
at Haskins Laboratories. Besides New Mexico, she has also taught at UCLA, the University of
Ottawa, and the Université Lumière Lyon . Her research has focused on prosody, particularly
timing and rhythm, in a variety of languages, especially French. This work has investigated
durational variation, vowel devoicing, and intonational patterns as evidence of French pros-
odic structure, and also considers listeners’ understanding of this structure.
L ORENZO S PREAFICO is a researcher at the Free University of Bozen-Bolzano. His current
research interests include monolingual, bilingual, and L phonological acquisition, with special
emphasis on articulatory phonetics. He has co-edited the volume Rhotics. New data and
perspectives (BUP).
P OLINA V ASILIEV is a Visiting Assistant Professor in the Department of Spanish and Portu-
guese at UCLA. Her main area of research is second-language speech perception, with a
particular interest in English-speaking learners of Spanish and Portuguese.
A LESSANDRO V IETTI is a tenured researcher in Linguistics at the Free University of Bozen-
Bolzano and Director of the ALPS (Alpine Laboratory of Phonetics Sciences). His research
fields are laboratory phonology and sociophonetics. He has written on phonetic variation in
bilingual speakers, combining sociolinguistic, articulatory, and acoustic data. He has recently
co-edited the volume Rhotics. New data and perspectives ().
OUP CORRECTED PROOF – FINAL, 12/10/2018, SPi
 OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

Romance sounds
New insights for old issues

M A R K G I B S O N A N D JU A N A G IL

. Why Romance, why now?

The study of Romance sounds, and their structure, has for centuries occupied a
foundational position in core phonetic and phonological research. And for good
reason. By examining the typological symmetries and asymmetries among the
Romance languages we have learned much about the universal properties of language
and the production/perception mechanisms that underlie sound change and devel-
opment. Additionally, the errors and direction of errors, as well as nonstandard
pronunciations, which speakers produce synchronically in one language give a
glimpse into the impetus for diachronic changes in the sound system of other related
languages within the same family, in our case, the family of Romance languages. All
of these factors conspire to provide a rich terrain in which to formulate and test new
hypotheses related to sound systems and in which to ground new areas of speech
motor research.
It is, therefore, of utmost interest to obtain an overview of the different processes
that have conspired to shape the different Romance sound systems in order to test
how different yet interconnected tendencies in a single phonological environment
may emerge. To exemplify the relation between synchrony and diachrony in the
Romance languages, consider the synchronic process of final consonant lenition in
Andalusian Spanish, as well as in certain Spanish dialects of Hispanic America and
Equatorial New Guinea. In these dialects, word-final /n/ may often become velarized
(Canfield , ; López Morales ; Lipski , ; Darias Concepción,
Ruisánchez Regalado, and Dohotaru ; Wireback ), /s/ and /θ/ are routinely
aspirated and/or deleted in all coda contexts (Alonso ; Salvador , ;
Alarcos Llorach ; Mondéjar ; Contreras Jurado ; Terrell ; Lipski
, , ; Romero ; Penny ; Gerfen ; O’Neill ; Kochetov
and Colantoni ), and /ɾ/ and /d/ are often precluded altogether (Chela-Flores
; Alvar [] ; Lipski ), as are other word-final consonants. Although

Romance Phonetics and Phonology. First edition. Mark Gibson and Juana Gil (eds.)
This chapter © Mark Gibson and Juana Gil . First published  by Oxford University Press.
OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

 Mark Gibson and Juana Gil

such processes are interesting peculiarities in their own right, the symmetry between
final consonant lenition in these Spanish dialects and the diachronic processes
that have contrived to shape the structure of codas in modern French, Portuguese,
Galician, Sardinian (R. Sampson ), Gascon (R. Sampson ), and Provençal
(Fernández González ) are indubitable.
In addition to studying crosslinguistic similarities among the Romance languages,
there is much to be gained by examining the asymmetries among them, especially
those cases that seem to present certain typological contradictions. One of the most
interesting innovations in the aforementioned Spanish dialects is the velarization of
word-final /n/ because of its close relation to the diachronic development of vowel
nasalization in French, Portuguese, Gascon, and Provençal. The suppression of the
tongue tip gesture for [n] (in /n/![ŋ]) represents an intermediate stage of lenition en
route toward total coda elimination (Lipski ), which fulfills assumptions regard-
ing crosslinguistic preferences for CV syllables. Omission of the final nasal gesture
means that the nasal quality of the word-final consonant is passed to the preceding
vowel, which, depending on one’s theoretical leanings, can be interpreted by way
either of feature spreading or of gestural mistiming. However prolific lenition
processes have been across the Romance languages, though, there are also attested
cases of nasal velar intrusion following nasal vowels in some dialects of Brazilian
Portuguese (Barlaz, Fu, Liang, Shosted, and Shutton ), which challenge the
directionality approach to sound change based on universal preferences for certain
syllable types. So, on one hand, we see diachronic cases of nasal elimination that are
typologically symmetric among certain Romance languages, leaving in their trajec-
tories the nasal vowels of many Romance vowel inventories as the nucleus of a
preferred CV syllable, but, on the other hand, we also see the insertion of a nasal velar
gesture following nasal vowels in dialects of the same languages that originally lost
final /n/ (and eventually the velarized nasal), contradicting so-called universal inclin-
ations for CV syllables. Although somewhat befuddling, such a scenario raises
interesting doubts about the universality and direction of sound change, as well as
the production/perception mechanisms that give rise to such changes.
An additional asymmetry that deserves special attention here owing to its prom-
inence in the literature is the case of vowel prosthesis before word-initial /sC/ clusters
in the Western Romance languages. As has been widely reported (see Alkire and
Rosen  for a good review), in most of the Western Romance languages a
prosthetic vowel aligned to the left margin of the sibilant, presumably in order to
resyllabify the word-initial /s/ as the coda of a preceding syllable, a process that dates
back to Popular Latin (Alkire and Rosen ). And while diachronic relics can be
found in many of these languages today, not all retain the synchronic alignment of
such a vowel: In French, for example, the insertion of an epenthetic vowel is no
longer productive, yet in Spanish (Harris ; Cressey ), Catalan (R. Sampson
), Galician (Colina ), certain dialects of Brazilian (Major ; Carlisle
, , , ; Renzi ), and European Portuguese (Fikkert and Freitas
), as well as in some Northern Italian dialects (Repetti ), epenthesis is still
very much a productive process. Paradoxically, although vowel epenthesis occurs in
Brazilian Portuguese and in certain contexts of European Portuguese, the deletion
 OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

Romance sounds 

of phonological vowels in European Portuguese preceding word-initial /sC/ clusters

is also attested: escola![ʃkɔlɐ]) (Miguel ; d’Andrade and Rodrigues ;
Mateus and d’Andrade ; Freitas and Rodrigues ; Henriques ). In
Romanian, as will be discussed in great detail in this volume (see Chitoran and
Marin, Chapter ; Marin, Chapter ), no prosthesis is attested either historically or
synchronically, and /sC/ and /sCC/ word-initial onsets are admissible, as in modern
French and Standard Italian.
The Romance languages, therefore, constitute a privileged framework for analyz-
ing common and divergent phonetic and phonological trends, and trends among
languages in general, both from the level of production and from that of perception
and acquisition. Although the treatment of the distinct phenomena depends on the
specific model in which the inquiry is couched in each of these fields, a notion readily
evident in the diverse and varied chapters of this current volume, the fact remains
that Romance phonetics and phonology are well poised to enjoy many more gener-
ations occupying their deserved position in the study of human speech sounds.

. New approaches to classic problems

Why, one must ask, if many of the topics taken up in this volume have been so
thoroughly scrutinized for centuries in Romance studies, must we detain ourselves
now to take yet another look at these recurring topics? The answer is clear, and
involves both internal (disciplinary) and external (extra-disciplinary) advances in
technology and in our thinking as scholars. Such advances have led to a profound
reassessment of many of the core questions present throughout the history of
phonetics and phonology since their inception as disciplines.
Perhaps the most fundamental question is precisely that which concerns the
relationship between phonetics and phonology as areas of scholarly attention.
Throughout the decades, there has been a systematic tendency toward the shifting
and blurring (two interdependent processes) of the boundaries between the two
disciplines (see, for example, Scobbie ). Not surprisingly, the location and the
very nature of the interface that relates them are still subjects of intense academic
debate. In recent years, the incorporation of new computational methods and the
ever-increasing use of novel technologies have contributed exponentially to this
blurring and shifting of the boundaries, with no lack of research on all sides of the
debate. Just as formal and generative computational approaches have been (and are)
informed by the technological advances of their day, ranging from early work on the
Turing machine to modern, powerful new computational methods using sophisti-
cated multi-paradigm numerical computing platforms, so now are new theoretical
foci being born from innovative multi and transdisciplinary approaches to the study
of human speech sounds. Many of these foci are informed by work in Gestalt-based
psychology, theories of dynamic patterning, and complex systems which have given
rise to phenomenological theories of speech that have broken ground in the way the
perception–production cycle and the dimensions of speech phenomena are envis-
aged. Work in psychology, statistical modeling, machine learning, cognitive science,
artificial intelligence, and philosophy, many times taking advantage of speech
OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

 Mark Gibson and Juana Gil

production and perception data, suggest—in some authors’ opinions—that the high-level
codification and organization of speech sounds (i.e., competence) are directly con-
strained by the low-level principles that regulate speech production and perception
(i.e., performance). In other words, cognitive constructs of speech sounds are informed
and constrained—according to proponents of these theories—by the body (at least the
mechanics of the vocal tract). Such a paradigm shift is reflected in novel phonological
theories such as Articulatory Phonology (Browman and Goldstein ), as well
as other laboratory-based approaches that envisage the high and low levels of
speech as dimensions of the same complex and dynamical system, as opposed to
two independent systems.
These relatively new approaches contemplate the low-level continuous variables of
speech (and language) as the result of multiple interacting, and at times conflicting,
influences involving physiological, motoric, psychological, and cognitive factors.
Such a focus seeks to resolve many of the unanswered questions related to Chom-
skyan modularity and posits speech and language not only as high-level cognitive
constructs, but importantly, cognitive constructs that are informed by, and manifest
themselves in, spatiotemporally coordinated actions. While such modularity is still
the impetus for groundbreaking discoveries in the computational processing of
speech sounds, evidenced by their great success in text-to-speech technologies and
voice recognition systems among many others, the embodied, dynamical approach
espoused in certain academic circles seeks to reconcile phonetic and phonological
phenomena with data in other cognitive domains, though at the cost of further
blurring the boundaries between linguistic competence and performance. All of
these theoretical paradigms in speech production, perception, and acquisition, no
matter what their underlying proclivities, firmly position modern phonetics and
phonology at the crossroads of science, industry, and the humanities. And while
the languages under study have changed little in the past century, the way they
are addressed and the tools researchers have at their disposal with which to examine
them have. Unsurprisingly, researchers in Romance phonetics and phonology have
been at the forefront of these novel, laboratory-based approaches to the field and
are naturally represented in this volume.
All of this notwithstanding, not an insignificant number of scholars consider that
“the best way to gain an understanding of the computational system of phonology is
to assume that the phonetic substance (say, the spectral properties of sound waves)
that lead to the construction of phonological entities (say, feature matrices) never
reflects how the phonological entities are treated by the computational system”
(Halle and Reiss : ). From this point of view, the phonology should not be
based on phonetic-level variables, since the phenomena that phonetics is supposed to
explain can be derived without any reference to phonetic substance. Thus, it is not
necessary to redundantly introduce the articulatory or acoustic principles that guide
the acquisition and the historical change of a language in an already constituted
synchronous grammar (see Newmeyer ; Hyman ; Barnes , among
others). Optimality Theory (Prince and Smolensky [] ), for example, in
its most classic conception, shares with the previous formalist models the goal of
formulating an explicit theory of linguistic competence, maintaining traditional
 OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

Romance sounds 

generative assumptions about phonology understood as a computational system

aimed at explaining the discrepancies between underlying and surface representa-
tions. The chapter by Jiménez, Lloret, and Pons-Moll included in the present volume
(Chapter ) subscribes precisely to this concept of grammar, endeavoring to evalu-
ate, from the analysis of the modifications that glides undergo in Spanish and Catalan
in certain contexts, the adequacy of the model in addressing the typological differ-
ences between the two languages and their varieties; in fact, the Optimality-Theoretic
approach presented seems to be capable of deriving the whole typology of outcomes
from the same constraint set. In the same vein, Martínez-Gil’s chapter (Chapter )
provides a strictly formal explanation of the distribution of vowels in Galician from
a more recent version of Optimality Theory known as Stratal-OT (see Bermúdez-
Otero, ), a modular approach to the phonology–morphology interface that
envisions multiple levels of organization for the phonological grammar in line with
Lexical Phonology (Mohanan ) and Lexical Morphology (Kiparsky a).
Specifically, Martínez-Gil’s study examines the upper vs. lower mid-vowel contrast
and the reduction process that these contrasts experience in unstressed positions.
The author concludes that the Stratal-OT model allows a simpler treatment of the
question than that of other more classic versions of Optimality Theory.
The analysis of the various phenomena documented in the Romance languages,
thus, serves to elucidate the more profound questions concerning what is perhaps the
most fundamental aspect of all those concerned with phonic studies, and which dates
back to Saussure: that is, the relationship between form and substance, and, ultim-
ately, the vision of linguistic knowledge as either internal, modular, mental, and
autonomous or as external, physical, and functional.
So, if we return to the Romance epenthetic vowel example mentioned in
Section ., from a formalist approach structural well-formedness of complex onset
sequences is determined by the linear arrangement of atemporal segments, which is
governed by rules or constraints that make some abstract reference to static vocal
tract states, and the sonority of the individual units and their distance from one
another in a sonority hierarchy (Harris ; Clements ; Colina ; Martínez-
Gil , ; Parker ; Wright ). Word-initial /sC/ clusters are considered
typologically marked because they do not adhere to the general sonority profiles in
onsets (see Eckman and Iverson ; Carlisle ). Harris (: –), addressing
the Spanish data, proposes that complex onset formation is governed by a constraint
requiring a minimal sonority distance of two¹ between the segments, while Parker
() proposes a minimal sonority distance of three for Spanish complex onsets.
While with certain contingencies both approaches are computationally feasible ways
to motivate epenthesis as a repair strategy in /sC/ sequences, the inherent cyclicity of
the explanation based on sonority distance raises doubts among many researchers

¹ Patterns of vocal tract constriction in complex onsets are projected onto the grammar as restrictions
that stipulate a minimal sonority distance between the segments. First, segments are collocated on a
sonority hierarchy on the basis of their respective manner specifications: Stops < Fricatives < Nasals <
Liquids < Glides < Vowels. Later, each hierarchical position is allocated an abstract numerical value:
() Stops < () Fricatives < () Nasals < () Liquids < () Glides < () Vowels.
OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

 Mark Gibson and Juana Gil

regarding the cognitive validity of such constraints and their capacity to explain
phonological patterning (see Ohala ). For this very reason, new approaches to
the phonology have tried to explain the phenomenon not only in alluding to
conditioning related to production mechanisms but by relating those mechanisms
of production to a perceptual impetus. This focus on the interaction between
production and perception represents one of the most interesting theoretical and
methodological innovations that have emerged in recent decades, and is ever present
in this volume.

. Advances in speech production: from articulation action

to acoustic consequences
According to Fant (), in summarizing the state of the fields of phonetics and
phonology in the second half of the twentieth century, variability and variation in
speech constitute “a basic theme” from the moment in which “invariance exists in a
relational sense only, to be tested ‘ceteris paribus’, that is in the same context.
Absolute invariance is a property of the perceptual-cognitive process induced by
linguistic competence rather than a property of the physical form” (p. ). Indeed,
the sources of both intra- and inter-speaker variability in speech are many and
diverse, and continue to be an indisputable challenge that must be addressed
in order to advance general knowledge in phonetics and phonology, and in the
development of any of their applications.
To meet this challenge of finding the source of variability, today researchers have
at their fingertips instrumentation that one hundred years ago would most certainly
have seemed like science fiction, and which affords the opportunity to empirically
address hypotheses related to production (and also to sound development and
perception) that were previously off limits, especially with regard to the temporal
dimension of speech, which until relatively recently was all but discarded in the
phonological literature. This, combined with the global trend toward multidisciplin-
ary approaches to scientific inquiry, has led many researchers to posit brand new
methodologies with which to test phonological hypotheses that incorporate to a
greater or lesser extent the use of the phonetic substance. This turn of events is
especially appreciable in the realm of speech production, particularly in studies
dealing with articulation. Several of these new protocols and instrumentation—
electropalatography (EPG), ultrasound imaging, and electromagnetic articulography
(EMA), among others—, which examine various types of data collected from both
spontaneous speech and laboratory corpora, are represented in the present volume.
With this in mind, Celata, Vietti, and Spreafico (Chapter ) start off the part on
articulatory studies offering an integrated analysis of rhotic variation in Italian using
a state-of-the-art, custom-designed, synchronized ultrasound imaging and EPG
system in order to characterize tongue tip and tongue body movement as well as to
register linguo-palatal contact. Among its many contributions, this study corrobor-
ates the staggering amount of variation that characterizes the production of rhotic
sounds crosslinguistically, a fact substantiated, also in this current volume, in Blecua
and Cicres’ (Chapter ) acoustic study of rhotic variation in Spanish.
 OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

Romance sounds 

Later, in Chapter , Chitoran and Marin offer an analysis of Romanian diphthongs

in hiatus sequences across word boundaries using kinematic (EMA) and acoustic
data. These authors provide evidence that Romanian vowels may form three distinct
categories of syllabic nuclei, and importantly, that each type of syllabic nucleus
involves different timing relations and degrees of coarticulation between the vowels.
In Chapter , Marin continues with an electromagnetic articulographic analysis
of four Romanian onset clusters (/spl-/, /spr-/, /skl-/, /skr-/). As the author eloquently
explains, three consonant clusters have been less systematically examined cross-
linguistically, and inconsistent results between languages render basing any deter-
mination of syllable affiliation on the c‑center hypothesis (see Browman and
Goldstein ) nettlesome. In her study of five native Romanian speakers, Marin
shows that although two-consonant clusters indeed show a c‑center effect, three-
consonant clusters do not.
To conclude this part on articulation, Recasens and Mira, in Chapter , also use an
experimental protocol combining electropalatographic and acoustic data in order to
explore the articulatory symmetry in lingual fricatives and affricates in Eastern,
Western, and Valencian Catalan. Generally, one can speak of the existence of
articulatory symmetry when the effects exerted on the articulation of certain seg-
ments by the different bases of articulation of the diverse dialects or languages are
observed equally in other elements of the system, though they might not share all the
specifications for place and manner of articulation. Effectively, Recasens and Mira
provide strong evidence for articulatory symmetry between fricative + fricative and
affricate + fricative sequences comprised of /s/ and /ʃ/ in the variants of Catalan
analyzed.
In spite of the diverse advances that have been made in articulatory research,
acoustic analysis remains the most accessible and widespread method for analyzing
speech production and its variability, especially due to the proliferation of free online
analysis software such as Praat (Boersma ) and many others. Precisely for this
reason, an entire part of the current volume is dedicated to acoustic studies. To
commence this part on acoustics, in Blecua and Cicres’ study of Spanish rhotics
(Chapter ) the authors detail the variation in Spanish rhotic production using
spontaneous speech, and address new contextual factors that condition this variation,
which have to date not been addressed in the literature. The authors show that the
more relaxed the speech style, the greater the degree of relaxation and weakening of
the various rhotic realizations, confirming results found in previous studies.
In Chapter , Regueira and Ginzo’s crosslinguistic socio-acoustic study of European
Portuguese and Galician voiceless sibilants details the acoustic variation, in this case of
sibilant production, between these two closely related languages, which may lead to
further change over time. Here, the authors found a great deal of both inter- and
intraspeaker variation for the realization of the alveolar sibilants, though this trend was
found most notably in Galician, Portuguese sibilants being generally more stable.
Perhaps, as Regueira and Ginzo point out, this greater variability of the Galician
sibilants is due to the fact that this language coexists with Spanish in the northern
region of Galicia, and that the relatively recent and incomplete diffusion of a Standard
Galician has still not achieved a sufficient degree of leveling.
OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

 Mark Gibson and Juana Gil

In the same vein, Gendrot, Adda-Decker, and Santiago (Chapter ) use acoustic
data to address vowel production in Spanish and French, and conclude that the
acoustic variations observed in the realization of the different vowels correspond to
different factors in both languages. For example, word frequency seems to play a role
in Spanish, such that higher-frequency words show less acoustic dispersion of vowel
formants, a result that does not hold for French. The vowel system of Spanish is
relatively simple, with only five vowel phonemes, whereas that of standard French
has thirteen, allowing for greater variability in the first language than in the second,
because the Spanish vowels maintain a greater acoustic distance. For that reason,
higher variation in Spanish does not jeopardize the vowels’ distinctiveness and the
consequent intelligibility of the utterance.
We see therefore in all of these studies that acoustic analysis, in effect, provides a
glimpse into how linguistic units, whether segments, syllables, or words, manifest
themselves physically with different properties depending on certain biological,
emotional, and physical characteristics of the speaker, and/or the different properties
related to the environment—linguistic, stylistic, social, geographical—, and allows
researchers to quantify such differences. Paraphrasing Ciocca and Whitehill (:
), the greatest advantage of this type of analysis is that it serves as a bridge
between articulatory and auditory information, and, in this sense, allows inferences
about the movements that take place in the vocal tract and their auditory repercus-
sion, thus helping us to understand the complex relationship between articulation
and perception. In Chapter  of the present volume, Campos-Astorkiza buttresses
this very notion, offering a model of voicing assimilation in preconsonantal sibilants
from a laboratory-based focus using acoustic data to propose that voice assimilation
is a byproduct of gestural mistiming between glottal and articulatory action that has a
linear effect in the acoustic domain.
Finally, acoustic analysis is considered to be an objective procedure less suscep-
tible to the subjectivity of the researcher. However, as is clear from many of the
chapters included in this volume (see, for example, Calamai, Chapter ; Madureira,
Chapter ; Racine and Detey, Chapter ), findings obtained from acoustic ana-
lyses need to be corroborated with perceptual testing in order to verify exactly
which acoustic dimensions are really relevant to the perception of human speech,
not to mention how the cues relating to the different acoustic dimensions are
weighted, and more generally, the nature of the relationship between production
and perception.

. Breakthroughs in speech perception

.. The application of perception in phonology
Not until relatively recently has perception begun to occupy a central place in
mainstream phonological research, precisely in the wake of the emergence of the
new theoretical approaches mentioned in the previous section, particularly experi-
mentally based theories of the phonology espousing a nonmodularized view of
phonetics and phonology that enables the incorporation of experimental techniques
to account for phonological processes. The development of more recent versions of
 OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

Romance sounds 

Optimality Theory has also allowed the inclusion of perceptual considerations in the
model, moving us closer to understanding the narrow relationship between produc-
tion and perception and the effects on the synchronous grammar. As follows,
perception has moved beyond the realm of psychology proper, and allowed phono-
logists to postulate groundbreaking theories regarding the processing and compre-
hension of speech sounds, and the interaction between audition, contextual
constraints on intelligibility, the role of memory, and the analysis of comparative
crosslinguistic perception, among countless other considerations (see Hume and
Johnson ; Hawkins , among many others).The fact that speech perception
has attracted the attention of phonologists in recent decades has made it possible to
explain certain processes that have previously only been approached from the point
of view of production. This paradigm shift, again, is manifest in the realm of
Romance studies, and many of the scholars included in this volume have been at
the forefront of the theoretical, experimental, and historical breakthroughs made
possible by this innovation.
To cite but one example of how the incorporation of perceptual considerations
into core phonological research has opened the metaphorical playing field for
phonologists, let us return to the synchronous cases of word-initial vowel epenthesis
in Spanish cited in Section .. Traditionally vowel epenthesis has been considered a
production phenomenon that results from a function of the productive Spanish
grammar, which references constraints that govern structural well-formedness
(although much debate exists regarding the underlying principles that regulate
onset formation). Besides the previously mentioned explanation based on sonority
categories, Gibson (b), taking inspiration from Davidson and Stone ()²
shows, in a gestural-based version of Optimality Theory (Gafos )—which has
subsequently become known as Gestural OT—how the ranking of certain constraints
referencing coordination relations between articulatory gestures can theoretically
drive the production of a vowel to the left side of the illicit /sC/ cluster without
resorting to constraints referencing sonority values. Underlying this proposal is the
idea that high-level information referencing syllable affiliation (structure) is
embodied in the spatiotemporal relations among articulatory gestures (task). How-
ever, the available articulatory data for Spanish and other Romance languages offer
inconsistent evidence for a timing-based explanation for vowel epenthesis. Further,
the problem in relating vowel insertion to gestural timing is that we still do not know

² In order to empirically test how nonphonological vocoids can emerge as an upshot of gestural
mistiming, Davidson and Stone () elicited non-native /sibilant+consonant/ tokens to native speakers
of English. The hypothesis was that if mistiming due to the modification of non-native coordination
relations were responsible for the insertion of an excrescent schwa, then the articulatory comparisons of the
schwa in pairs such as [zəɡ] (from the Polish nonce zgama) and [sək] (from the English succumb) should
exhibit important differences. If, however, the articulatory comparisons revealed no significant differences
in tongue posture between the two schwas, then it was hypothesized that the insertion of the schwa in the
test tokens was phonological, and not related to the mistiming of the non-native target. Although results
were highly speaker-dependent, this study showed significant differences in tongue posture for the two
types of vocoids, hinting at the possibility that vowel insertion may result from the mistiming of vocal tract
action, and that the extra-lexical vocoid-like element may not be a discrete phonological unit.
OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

 Mark Gibson and Juana Gil

enough about why typological differences exist between languages with regard to
timing nor why languages that exhibit typological similarities related to the temporal
organization of the gestures in complex onset clusters exhibit different strategies
to repair illicit gestural parses and vice versa. Although this can be explained by
language-specific (or even dialect-specific) phonotactic constraints based on spatio-
temporal coordination, the explanatory power to justify the emergence of a vowel to
the left side of the /sC/ cluster is crucially challenged in the sense that it still does not
explain why insertion emerges in one language but not in another, nor the conditions
on which ill formation of the /sC/ onset are based, not to mention the specific vowel
that inserts across the different Romance languages and dialects.
In the face of these problems and a myriad of others, a growing number of studies
provide evidence to suggest that vowel prosthesis may be a perceptual phenomenon
based on native-language phonological biases. The idea that listeners form biases
toward certain aspects of their native language is robustly supported in Bertoncini
and Mehler (), Hillenbrand (), Werker and Tees (), and Jusczyk,
Friederici, Wessels, Svenkerud, and Jusczyk (). The claim that the perception
of a continuous acoustic stream can be distorted by the phonology of a listener/
perceiver was first introduced in Hallé, Segui, Frauenfelder, and Meunier () and
elaborated on in Dupoux, Hirose, Kakehi, Pallier, and Mehler (), Dehaene-
Lambertz, Dupoux, and Gout (), Kabak and Idsardi (). In these studies it
was shown that an illusory segment can be perceived even though no acoustic or
auditory evidence exists for such a segment.
More recently, following Hallé et al. (), Gibson (a) performed a series of
identification and discrimination tests with fifty native Spanish-speaking subjects
aged ten to eleven. In the first test, the subjects were played nonce stimuli containing
both licit and illicit word-initial onset clusters. The subjects were asked to respond
with the number of syllables they heard. The results indicate that the perception of an
illusory syllable (i) is prevalent in all target sequences, and (ii) is unsupported in non-
target stimuli. A set of follow-up tasks revealed that the perception of an extra syllable
resulted from an erroneous interpretation of the word-initial /sC/ sequence. Import-
antly, the data from the tests reveal that the fundamental conditions of the bias
motivating the prosthetic vowel go beyond the mere linear organization of spatial
characteristics.³
To cite another example of the role perception may play in explaining phono-
logical structure, Silvia Calamai, in Chapter  of this volume, offers experimental
perceptual data in order to understand the nature of an Italian phenomenon known
as anaphonesis, further linking synchronic grammars to diachronic changes. In the
historically attested underlying vowel inventory of Italian, the Latin short vowels
ĭ and ǔ opened to [e] and [o], but not all Italian dialects exhibit such behavior,

³ Additionally, and perhaps just as interesting, is the difference in vowels that insert. In Catalan
(Wheeler ), Galician (Colina ), and Spanish an [e] inserts (Cressey ), while epenthesizing
zones in Brazil and Italy insert [e]–[ɪ]/[i] respectively (Major ; Carlisle , , , ; Renzi
). In French, [i] inserted originally, but was lowered to [e] when Romance vowel lowering took effect
(R. Sampson ).
 OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

Romance sounds 

especially those dialects in which the vowel occurs before a velar nasal consonant:
VĬNCO > vinco, *venco; LĬNGUA > lingua, *lengua. This phenomenon (so-called
anaphonesis of the second type) may equally emerge either by way of phonological
innovation or as a conservative strategy (known in other circles as faithfulness) to
maintain the high vowel. The first hypothesis, based on phonological innovation, is
centered around the notion that the weakness of the nasal velar and the maximum
anticipatory nasalization of the vowel in this context originate at a given moment,
such that the latter is lengthened and becomes more tense, and thus, closes. The
second hypothesis, however, posits that the aperture does not take place simply
because of nasalization and the sonority of the postnasal consonant. The result of
Calamai’s perception experiment seems to point more toward phonological conser-
vatism, and not innovation, to explain this phenomenon.
This process of anaphonesis, which does not affect anterior and posterior vowels
alike, is much more complex than has been summarized here, but at the heart of this
question lies the idea that the essence of the diachronic phenomenon of anaphonesis
can be corroborated experimentally by synchronic perceptual behavior, as Calamai
proposes, based on the assumption that changes in the degree of nasal vowel opening
originate in the listener, who attributes some of the acoustic consequences of
nasalization to a distinct articulation of the vowel. In this sense, Calamai’s underlying
focus is reminiscent of Ohala ()—who, boldly yet assuredly, surmises that
coarticulatory changes always have a very important perceptual component—and
further proposals made in Evolutionary Phonology (Blevins , ), which argue
that the phonetic-phonological patterns to which speakers and listeners adjust to in
synchrony can explain attested changes over time.
The cases of prosthetic [e] insertion and Italian anaphonesis corroborate the
relevance of perception to phonological patterning and distribution, which until
relatively recently could only be addressed theoretically, and from the perspective
of production. Following in this line, Smith, in Chapter  of this volume, analyzes
the phrasal units in French prosody (accentual phrase, intonational phrase, inter-
mediate phrase) from a perceptual perspective and demonstrates the way in which
French-speaking listeners perceive certain prosodic structures—dislocations, wide
and narrow focus, and particularly long noun phrases that function as subject—as
well as how these structures align with the phrasal units. Crucially, Smith provides an
analysis of how speakers interpret these structures and how they use them in their
communicative exchanges (see, for example, Local ), which would not be
possible exclusively from the standpoint of production.

.. Perceiving fine phonetic detail

In recent decades a relatively new movement toward uncovering the role fine-grained
phonetic detail plays in speech encoding and processing has emerged in order to
address how different acoustic cues and dimensions are relevant to the perception of
human speech. The impact of this trend is appreciable in an ever-growing number of
new academic fields such as forensic phonetics, among many others (see Fernández
Trinidad and Rojo Abuin, chapter ). Additionally, this movement has also found a
niche in the industrial sector, where the inclusion of more fine-grained phonetic
OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

 Mark Gibson and Juana Gil

detail is employed to enhance voice recognition and identification systems. Nguyen

(), citing Hawkins (), defines phonetic detail as “subphonemic phonetic
variability that contributes to phonological or other contrasts that distinguish mean-
ings, but not necessarily lexical items” (p. ). And although a number of studies in
speech perception have shown that speakers utilize these details to process and
understand the message encoded in speech (cf. Hawkins , , among many
others mentionable), these oftentimes microscopic details are merely skimmed over,
if not ignored, in most classical theoretical frameworks. Notwithstanding, the role
that they play in the listening to and understanding of speech has gained much
currency in recent decades, in part owing to the fast-growing number of spontaneous
speech databases that are available for analysis, but also thanks to new conceptions of
the phonology that envisage the phonological representation as a detailed and
complex construct that possesses a high grade of indexical and phonetic information
(Pierrehumbert ). As a result, such non-abstractionist models are not hamstrung
by the necessity to eliminate redundancies (for a detailed discussion on exemplar
models of phonological encoding, see for example Gahl and Yu ).
Evidence for language-specific phonetic detail in the encoding of phonological
categories without a doubt obliges us to rethink the relationship between phonetics
and phonology. It should be understood that the analysis of such phonetic detail is
not confined to those subphonemic properties or peculiarities that would be associ-
ated with the lexical representation (for example, the exact point of articulation of a
segment, the varying extent of coarticulation or variable duration of voice-onset
time), but also encompasses many other far-reaching features that contribute equally
to the correct understanding and interpretation of statements that are nested in
communicative interactions, such as characteristics related to language-specific
articulatory settings (see Recasens and Mira, Chapter , in this volume for further
details on language specific articulatory settings), voice quality, or certain micro-
fluctuations in tone. As a consequence of the varied nature of the types of phonetic
detail, they can be classified, adapting a similar assertion by Payne (b), as either
local-domain or global-domain. In the former, the variability that characterizes the
realization of the same phoneme or a paradigmatic contrast—in a single language or
in different languages—would be understood as derived from more or less subtle
short-range differentiating phonetic details. By contrast, the latter refers to the
variability associated with differences extending beyond the segment, which is due
in many cases to distinctions of long-range phonetic detail.
A fine example demonstrating the relevance of local-domain phonetic detail in
speech perception is Llisterri and Schwab’s chapter in this volume (Chapter ),
which offers a crosslinguistic perceptual analysis of Spanish lexical stress by French
listeners (fixed stress in French, relatively free in Spanish; demarcating stress in
French, distinctive stress in Spanish, cf. also Gendrot et al., Chapter ), and of their
sensibility to minor variations of duration, intensity, and f₀. The perceptual cues that
listeners identify with prominence are not the same in both languages, and the
experiments carried out by Llisterri and Schwab show that “French listeners seem
to be more sensitive to fine-grained details than Spanish listeners, whether in a task
that implies an acoustic processing of lexical stress or in a more demanding task that
 OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

Romance sounds 

implies not only an acoustic but also a lexical processing of stress” (p. ). In
particular, it was found that French-speaking listeners who have no knowledge of
Spanish are sensitive to different cues, and probably assign different cue weights in
relation to the acoustic dimensions that are crucial in their language. Only as their
phonological and lexical representations of the second language (Spanish) are
updated will the French listeners be sensitive to the relative perceptual cues.
With regard to global-domain phonetic detail, Fernández Trinidad and Rojo
Abuin (Chapter ), in their study of the perception of voice quality in Italian, and
Madureira (Chapter ), who deals with the production and perception of emotional
substance transmitted by way of rhotic variation and voice quality settings in
Brazilian Portuguese, demonstrate how meaning can be manipulated, either by desire
to transmit some affective state of the speaker, as is the case in Chapter , or
purposefully to disguise one’s identity, as is the case in Chapter , by making fine-
grained adjustments to the acoustic signal. It has indeed been found in various
languages that listeners are sensitive to more or less subtle changes in voice quality
and that such alterations can provide valuable information for the listener.⁴ Thus,
establishing the role that subtle changes in phonation and other fine-grained phon-
etic details have in the sound-to-sense (and vice versa) mapping in the Romance
languages has very interesting implications vis-à-vis phonetic universals and typo-
logical symmetries. The chapters dealing with these topics in this volume certainly
advance knowledge in that direction.
In short, a better understanding of the perception of speech and its transcendence
for communication has become an object of privileged study in recent times.
Moreover, in addition to the patent theoretical advances this research inspires, an
enriched understanding of how humans perceive speech has both academic and
practical benefits in the domain of (first and second) language acquisition and
teaching, given perception’s link to sound discrimination and classification, neces-
sary initial steps for a child or an adult to build their early lexicon.

. Discoveries in language acquisition and learning

Perceptual issues, along with production, occupy a prominent place in studies
dealing with second-language acquisition, in both monolingual and bilingual set-
tings. It is well known now that the perception of L sounds is mediated by the
phonic categories existing in the L, which in turn conditions the production of
second language sounds (Best ; Flege ; Best and Tyler ; van Leussen and
Escudero ). Llisterri (), in a frequently cited study addressing the link
between perception and production in L, concluded that researchers are still far
from knowing the exact way in which both modalities, perception and production, of
an L are interrelated since perceptual capacity cannot always predict articulatory

⁴ For example, creaky voice, a nonmodal phonation mode, has been purported to have linguistic-
phonetic value in Finnish, as a component of the turntaking system, independently of other syntactic or
intonative resources (Ogden , ). A similar phenomenon has been found for English as well (Laver
: ).
OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

 Mark Gibson and Juana Gil

accuracy and articulatory accuracy does not always reflect perceptual capacity. The
critical review that Elvin, Vasiliev, and Escudero offer in Chapter  of the present
volume regarding the perception and production of Spanish and Portuguese vowels
arrives at the same conclusion: perception and production maintain a close and
undoubted relationship, although it is not possible to determine yet if perceptual
development precedes production or vice versa.
Later, Simonet, in Chapter  of this volume, expounds the results of experiments
related to perception and production in bilingual Spanish–Catalan contexts, citing
potential effects of early linguistic experience on the perception and production of
sounds in bilingual subjects. In light of the previous research that Simonet reviews,
it seems reasonable to claim that both the ability to perceive, process, and represent
contrasts and phonological traits and the ability to produce them are conditioned,
at least to some extent, by the specific sequential order of acquisition of the two
languages. As Simonet explains, if there is not massive exposure to the second
language during the first year(s) of life, the child is unlikely to achieve competence
on a par with native speakers. Notwithstanding, with regard to production, crude
explanations do not fit, because many noncontrollable factors such as linguistic
experience, social networks, and so forth may condition a speaker’s behavior, leading
to a perceivable non-native-like accent (Piske, MacKay, and Flege ).
One of the noncontrollable factors that intervenes in second-language learning is
the greater or lesser perceived proximity of the L and L in question, predicting a
greater or lesser probability of achieving native-like production (Elvin, Escudero, and
Vasiliev ). Such proximity, as Bosch elaborates on in Chapter  of the current
volume, also plays an important role in simultaneous bilingual language acquisition.
In her chapter, Bosch reviews data related to early language differentiation skills,
phonetic perception, and word segmentation abilities from infants growing up in
Catalan–Spanish contexts and compares them with data obtained from children
who are exposed to more distant languages. The assumption is that greater or lesser
proximity between L and L may explain many differences in the acquisition
process of the different groups of bilingual children, in some cases hindering certain
aspects of acquisition while at the same time facilitating others. The author explains
that at the initial state of bilingual acquisition, the perception by infants of cross-
linguistic closeness or distance must come from the broad phonic configuration of
the languages, especially those pertaining to prosody (i.e., rhythm and intonation),
and also—probably, although little is known in this respect—from external or
internal articulatory settings, which are linked to the language-specific configuration
of the vocal apparatus (Laver ).⁵ In this sense, broad classifications, such as

⁵ As is well known, some articulatory settings are internal (not externalized, for example, the permanent
nasalization of certain sounds), but others are external, that is to say that they are visibly perceptible and
recoverable (for example, an articulatory setting consisting in moving the lips in an exaggerated motion
during infant-directed speech, which gives rise to a type of speech with very definite perceptual charac-
teristics, see Green, Nip, Wilson, Mefferd, and Yunusova ). It would be interesting to explore whether
the acoustic cues that result from these settings, which represent microscopic phonetic detail that the
infants receive from the ambient input, are also used, as prosody is, to discriminate languages.
 OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

Romance sounds 

Romance languages versus Germanic languages, do not capture the nuances that
generate the impression of phonetic, phonological, and prosodic crosslinguistic
proximity and distance.
So, the notion of language-specific phonetic detail emerges again as an important
factor in the field of language acquisition and learning. Selective attention to a certain
kind of phonetic detail specific to the L, rather than to other relevant cues in the L,
has consistently been found to account for the misperception of new L contrasts,
leading to variable degrees of non-native accentedness (Kondaurova and Francis
). Phonetic detail has long remained outside the grammar, but the point is
that these microscopic cues are not always automatic, nor mere repercussions of
random movements in the vocal tract, but rather constitute systematic codes used by
speakers to transmit extra and paralinguistic information in a particular language. So
while they might reside outside the grammar, they are not outside the language.
Thus, one of the tasks that L learners must tackle is assimilating and internalizing
these gradual, quantitative, and subphonemic patterns of detail, as well as the
distribution of said variability of the patterns, in order to form more robust phono-
logical and lexical abstractions that can later be used to extend generalizations to new
words and utterances.
Nevertheless, the quantitative and qualitative study of speech variability that
precedes any subsequent analysis of its details can be carried out only if abundant
recorded data of spontaneous speech, that is, a corpus, are readily available. As
G. Sampson () points out, “the recent rise in corpus-based research methods
has been at least partly due to a reaction against . . . unempirical style of linguistic
research” (p. ). Concretely, for phonology this implies an ever-tightening rela-
tionship with innovative theoretical models of a bottom-up nature (from the low-
level dimension of speech to the high-level informational dimension) that utilize
instrumental and experimental procedures to falsify or confirm their claims.⁶
In light of this new technology-driven focus in all strata of the speech sciences,
Rojo (), citing Dyson (), remarks that the main difference between concep-
tual revolutions and tool-driven revolutions is that, while conceptual revolutions
involve substituting one scientific paradigm for another in order to explain unre-
solved issues in new ways, tool-driven revolutions involve discovering brand new
issues to explain. In phonic studies, speech technologies and computer science in
general have made it possible to collect enormous quantities of spoken registers for
varying purposes, analyze them later statistically, and extract new insights and
hypotheses regarding the production, perception, processing, and computation of
speech sounds in any number of situations, constituting a veritable revolution in all
fields involved. The natural upshot of such a revolution in speech technologies is that
the study of speech sounds and systems, with special reference to the Romance

⁶ It is important to point out that the expressions used in Spanish, fonología de corpus, or in French,
phonologie de corpus, to translate corpus phonology are not especially adequate, since this framework does
not propose a different type of phonology, but rather combines the methods and theoretical principles of
phonology as we understand them with the support of a different type of instrumentation and analysis:
perhaps it would be more appropriate to speak of fonología con corpus / phonologie avec corpus.
OUP CORRECTED PROOF – FINAL, 3/10/2018, SPi

 Mark Gibson and Juana Gil

languages, is firmly positioned to occupy a permanent place in the burgeoning new

field known as Big Data.
Certainly, this technological revolution has also had a bearing on the current
volume, specifically Racine and Detey’s (Chapter ) corpus-based study in second
language acquisition (SLA). As Gut () explains, “the use of corpus-based
methods in SLA research, however, is still rare. Only a handful of studies have so
far tested explanatory SLA hypotheses against corpus data and have attempted to
validate previous findings obtained in small-scale studies with corpus analyses. Yet
these first results have been interpreted as very promising, so that many researchers
currently argue for a greater use and an increased development of second language
corpora for the analysis of second language data. . . . This is especially true for the area
of second language phonology, where corpus-based approaches are still few and far
between” (p. ). Responding to this need, Racine and Detey’s study examines the
phonemic contrast between the two French closed rounded vowels /y/ and /u/ by L
Spanish university students within the framework of the L French phonology
research program InterPhonologie du Français Contemporain (IPFC) (Detey and
Racine ; Racine and Detey ). The production of the vowels in question by
the Spanish speakers (learning French in Madrid, Spain, and Geneva, Switzerland) is
first perceptually evaluated by naive native French speakers, then acoustically ana-
lyzed, and, finally, examined using the coding procedure of the IPFC project, with
convergent results.
An observation that is discussed in the chapter and which is interesting to mention
here, again, is the importance of phonetic detail in understanding issues related to L
acquisition. As Racine and Detey persuasively explain in their chapter, the results
seem to suggest that the Madrid learners are aware that all French vowels are more
closed than the Spanish vowels and, as a result, exaggerate lip movement in the case
of /u/, leading to a non-native-sounding accentedness. This is a case, therefore, in
which the learner enhances a primary acoustic cue of the segment beyond what is
necessary. In the case of /y/, on the contrary, the students also accentuate its lower
degree of openness in relation to /u/, but do not sufficiently enhance its anteriority in
the horizontal plane. The mastery of these subtle differences is what eventually makes
a speaker be perceived as more native-like, even though they may not significantly
affect the intelligibility of the production. It is clear, therefore, that this type of finding
regarding the fine-grained phonetic detail, as well as its perceptual impact (with or
without phonological significance), which was made possible through the growing
presence of new speech technologies and their applications, is of considerable
importance for teaching the pronunciation of second languages.

. Aims, scope, and organization of the current volume

As can be deduced from the previous sections, the general proposal of the present
volume is to highlight innovative research in the classical and new areas of inquiry
outlined previously by outstanding senior researchers who have become influential
international competitors in their respective fields, and junior researchers who show
brilliant promise of becoming equally competitive contenders in the international
Another random document with
no related content on Scribd:
Aurora and stars, shed their wonder-light over the scenery, the boys were
once more happy and gay.
On the days—strange to say days when all was night—when the
temperature fell to 20° and 30° below zero, cold was not complained of, but
zero itself, with the wind-fiend raging, was misery that cannot be described.
Dr. Wright did everything a brave doctor could do to keep his people in
health and fit. Curtis was no longer commander save in name. He had to
cave in to the doctor, and do all he was bidden.
MacDonald told his queerest stories after dinner, and sang his love lilts
as heartsomely as do the blackbirds in early spring.
Everybody had come to look upon Mac as a brick, and his cheerful
Doric voice even in the dark was delightful to listen to. He used to “bag the
boys” at night, as he termed it, Charlie with Nick, and Walter with Nora.
“Bag them” snugly, too. He was like a mother to them. Of course all hands
turned in very early, and as Curtis’s bag (and Collie’s) and also Dr. Wright’s
were close to Mac’s and the boys’, the Yak-dogs filling up the intervals or
lying round the sides, Mac could lie and yarn, or even sing, to all hands for
two hours at a stretch. The British sailors were not far away in their bags,
and they could listen too.
There is no seaman in the world like our handy man the British, and
through all that long and trying Antarctic night these good fellows, though I
have said little about them, behaved like heroes.
All kinds of games could still be carried on in the light, but sleighing
was discontinued.
In these regions it is just after turning in that one feels most cold, but any
such course as warm drinks or nightcaps (drinkable, I mean) would make
matters worse.
Slap-dash and his people used often to worship the moon, just as they
had the sun. The sun may be the god of these poor souls, but the moon is his
high priest, and the Aurora are his angels.
Well, a religion of any sort is better than none.
Once when the moon was about three days old she took on a strange but
most lovely appearance. The stars, except the highest, which were
exceedingly brilliant, burned somewhat less brightly at the time. But it was
towards the moon all eyes turned.
 It was, if I may so describe it, a kind of rainbow moon. The outer arc
was of the deepest orange colour, the next and largest arc was pale yellow,
but brilliant, then an arc of radiant sea-green, while inside all was an arc of
pale but indescribably beautiful mauve.
Hitherto the boys and Ingomar himself had believed, or been taught to
believe, that the Aurora Borealis, or Northern Lights, with their fringe-like
bands of opal, pink, or green, were far more lovely than the Southern
magnetic lights, the Aurora Australis.
During their sojourn in the Antarctic they had time to alter their opinion.
I feel it is presumption on my part to attempt to describe a display of this
Aurora, because I shall hardly succeed in making myself understood.
Just imagine, if you can, a wide and wondrous arch, stretching from east
to west, and nearly halfway up the sky, more rounded than a rainbow, its
ends apparently within a few feet of the snow-field.
At first the arch resembled a vast chain, every link of which was a ring
of brightest gold, each link overlapping its neighbour to about one-half its
extent, but all turbulent, all a-quiver! But lo! as one gazed on it, strangely
fascinated, the rings, though still linked together, turned half-edge-on
towards the right. Then from each ring, as a spherical base, was suddenly
thrown out a triangle of glittering, darting, quivering, golden light.
But speedily is the apex of each triangle extended zenithwards, and
broadened out, till it resembles a brush. The rings get smaller and smaller
beneath, until they are but bright points of light like heads of comets; in
very truth, there is now a broad archway of comets, heads downward
towards the snow.
But listen. While the heads of these comets retain the brightness of stars
of gold, the extended brushes, or tails, are now bunches of rainbow-
coloured, flickering, dancing, darting light.
It is a bewildering sight, and it is hard to believe it real.
Gradually the tails get shorter, become once more the apexes of spherical
triangles, and dance, and disappear, the chain of golden rings becoming
once more visible as before.
All beneath this archway is a dark-blue sky, in which stars shine, and the
rest of the firmament is quite unaffected, though the mountains and snow-
clad valley borrow the colour and add to the bewildering grandeur of the
most marvellous transformation scene the world can ever witness.
I fear I have failed to give my youthful readers an adequate conception
of the Aurora. I feared I should fail before I commenced. But Britons—and
I am one—should never funk, and I have done my best.
* * * * *
It is strange, and sadly strange, that, although Dr. Wright and his men
had borne bravely up, throughout the livelong night of the dreary Antarctic
continent, as soon as day returned, revealing blue and ghastly faces,
sickness came.
This is no place in which to inquire into the cause of this sickness;
suffice it to say that it came, and the men, hitherto brave and hearty, began
to droop and shiver.
An optimist at most times, and ever ready to look upon the bright side of
circumstances, the doctor himself began now to fear the worst.
Long before my own experiences of Arctic life, there used to be in Polar
regions a disease called the black death.
Whether or not the illness that now attacked this little camp of heroes
was a species of that ailment, I am not prepared to say.
I hate to have too much gloom in my stories, or I could describe the
symptoms so graphically that you would shudder.
Suffice it to know that, though there were no unsightly swellings, and
though the faces of the sufferers retained even their complacency when fits
of shivering and cramp abated, they were melancholy and sad sights until
they either recovered or died.
Let me say at once that though both Charlie and Walter were ill a few
days, owing to the resiliency of youth they were not stricken down, and
speedily recovered so far as to be able to assist the truly sick.
It need not be said that Dr. Wright did all that any medical man could
have done. Just one or two of the Eskimos collapsed utterly, and died on the
third day. They were buried not far off in the snow. Two days after a sailor
followed them to the snow-field. He did not say much, even at the worst,
and finally he simply fell asleep. Only one out of the four other men
attacked recovered, and this was far more from good management and the
kindly nursing of Sheelah and Taffy than from medicine. In fact, though
wine did good when the patient was at the lowest ebb, and helped him to
fight his way round the corner to restoration, medicine was for the most part
useless.
Curtis was early down, and, strangely enough, considering how truly
brave he was, his spirits drooped to zero, and he gave up hope of himself
from the first.
Ingomar nursed his dear friend indefatigably, and when, overcome with
fatigue, he dropped off to sleep, either Sheelah or Taffy was always sitting
by his brother when he awoke.
I cannot really testify in strong enough language to the marvellous
qualities of those gentle little Yak women as sick nurses.
We may laugh at such people, ah! curious though their customs be, and
droll their manners, they are our sisters before God.
Slap-dash remained his old self.
Let me cut this all short by saying that of all the crew of brave men, only
twelve remained to take the road back to the seashore.
Perhaps as sad a case as any was that of poor MacDonald, who had been
so long the life and soul of all the camp.
When Dr. Wright told the boys that he could only last a few hours, and
that they must go and see him now, they summoned courage enough to have
the interview.
They behaved splendidly in his presence, but as soon as they went into
the open air again they both utterly broke down and wept, until their hearts
appeared almost bursting.
“It does seem hard, does it not, Walter?” Charlie managed to say.
“Always so kind and good,” said Walter.
“Ay, ay, and I never knew I loved him half so much till now.”
* * * * *
Mac, once the hardy, resolute Scot, passed away that same day.
In the semi-darkness of the cave Ingomar was kneeling by his side and
holding his hand.
He had lived a Scot; he died a Scot.
Ingomar thought he had fallen into a slumber, so quiet did he lie. But he
spoke at last, though with feeble, faltering voice.
 “It’s you, isn’t it, Ingomar?”
“I’m here, dear Mac.”
“Well, I—I know I’m dying. I wouldn’t care—but mother——”
“What can I do to ease your mind?”
“She kens I love her—I’ve been single for her sake. Promise to get all
I’ve saved, Ingomar. Her dear auld-farrent[G] letters and my bank-book are
a’ in my box. Ingomar—you—promise?”
“Most sacredly.”
“God love you! She’ll no be lang ahint (behind) her laddie.”
He lay still a little while, and he spoke but once again—repeating a verse
of the 23rd Psalm.

“ ‘Yea, though I walk thro’ death’s dark veil,

Yet will I fear none ill;
For Thou art with me;
And Thy rod and staff me comfort still.’

“Is that the Aurora? Ingomar, tell me. Oh, how bright and how—joyful
—Father——”
He was gone!
He had seen the Aurora; but it was the morning dawn of a happier life.

CHAPTER XI

“ENGLAND, HOME, AND BEAUTY”

Six weeks after this, and when the captain of the Walrus had given the
explorers up for lost, after searching the snows in vain, for winter storms
had obliterated every track, ten men with two dog-sledges suddenly
appeared above Glen Bell on the ridge of the great tableland.
They rested there.
They knew they were seen.
In the stillness of the early summer’s morning they could hear the wild
shouts of greeting that arose from their shipmates.
 And you may easily guess that assistance was speedily on its way to the
top of the valley.
I leave you to guess also the kind of welcome accorded to men and dogs.
Why, Slap-dash himself came in for hugging, and Wallace hugged every
one indiscriminately all round. Dr. Wright, Ingomar, Curtis, and the boys
were all sadly worn and sallow. They had but little life in them. Even their
courage appeared to have left them. They smiled, it is true, but it was the
smile of sickly old men.
When they were helped on board at last, and had a little food and wine,
they begged for a bath and to be shaved. After this, and dressed in fresh
clothing, they were in some measure restored.
Captain Bell and the other officers of the Sea Elephant had come on
board, and to them Ingomar, who was stronger than the rest, told the sad
story of their terrible hardships, and their struggle to reach the ships. When
he spoke of poor MacDonald, there was not an eye in the room that was not
dimmed with tears.
But there! I myself must pull up. I would not have my very last chapter
dimmed with sorrow.
Suffice it to say that not only these five real heroes, but the Yak-Yaks,
including Slap-dash and Sheelah and Taffy, were in a month’s time their old
selves again.
Of the animals, strange to say the Shetland ponies, Jack and Gill, had
been least affected, while Wallace had returned hungry, Nick and Nora
standing by delightedly as he ate the food prepared for him. The
Newfoundland, as soon as he had finished, proposed a romp round the
decks. Wallace tried, but soon lay down to rest and pant.
“Another day, I hope,” the honest Collie appealed, “but somehow I feel a
little tired.”
When the good old Walrus was sawn out of her quarters and got into
blue water again, with all and everything on board, and when the Sea
Elephant lay quiet and still on the calm blue sea, a dinner was given on
board the flagship.
The blessing asked by good Captain Walt was a prayer of thanks to the
Almighty Power that had guided them through their trials, through sickness,
danger, and difficulty.
 I think all hands, fore and aft, who partook in the festivities, were just a
little great-hearted at first, but all sadness was soon dispelled.
They had all done their duties bravely and well, as British and American
sailors and soldiers always do.
So upon the whole a very happy evening was spent, the thought that next
day they would bear up once more for the shores of Merrie England—
England, home, and beauty—put life and spirit in them, and they retired at
last, happy and hopeful.
I don’t think that any one on board the Walrus or Sea Elephant is ever
likely to forget the sweetness of that Antarctic summer morning—the
morning of the start. The sea with its beauty-tints of opal and blue, a sea
studded with the snow-white of tiny bergs, the great mountains towering
skywards, and the world, the marvellous world, of bird-life.
Do you know that, great though their sufferings and hardships had been,
every one looked back to the scene of their adventures with just a little
feeling akin to sorrow!
Up steam!
Round go the screws, churning up a frothy white wake, slowly move the
ships away, slowly, and apparently reluctantly.
But, in a few hours’ time, those sturdy ships are merrily bobbing and
curtseying to each advancing wave, as if they really know that, at long, long
last, they are homeward bound.
And now nothing reigns aboard, fore or aft, except happiness and
general jollity, in which even the dogs themselves take part.
Homeward bound! Hurrah!
* * * * *
When, in about two months’ time, the Walrus and Sea Elephant came
quietly to anchor inside the breakwater of Plymouth, people gazed and
wondered what these two strange ships could be.
But when the truth was rumoured abroad that they were the Antarctic
voyagers, the wild welcome they received was enough to have turned the
heads of any sailors on this earth.
* * * * *
Parting!
 Yes, parting, yet parting—every one assured his shipmates—to meet
again and talk over old times.
The boys, Charlie and Walter, going off to their “uncle’s” home.
Dr. Wright to duties elsewhere.
The scientists to London.
Slap-dash and his dogs and Yak-Yaks, including faithful Sheelah and
Taffy, to London, with the scientists.
The boys got all the three dogs, and happy enough the dear fellows
seemed to get on shore again.
Parting! Ah, yes, it is a sad word, and so I leave it.
* * * * *
Ingomar, the prodigal son, returned to his home.
“Can you forgive me now, father?” he said, after he had embraced his
mother and sister.
“Bosh, boy!” cried the old man. “Go and sit down.” But there were tears
in his eyes nevertheless.
Curtis was here, too.
Curtis came home to find he had succeeded to a baronetcy and another
large estate. But this would not have stirred his spirits in the least had not
Marie greeted him so joyously.
He used to call her his Marie. In six weeks’ time she was his Marie in
reality.
They were married.
Ingomar says he will never marry. I simply smile.
He is owner, anyhow, of one of the most splendid yachts ever built in
America or England.
No ’long-shore yacht. Not built for racing or speed, but comfort,
pleasure, and beauty. Curtis has left the service. The yacht takes very long,
delightful cruises, but wherever she goes with Ingomar, her master, both
Arnold Curtis and his sweet wife go along as well.
My story is ended, my tale is told. I have only to say “Good-bye, my boy
reader, and God be with us all.”
I trust and hope we’ll meet again another day.
 FOOTNOTES:
[A] Skis, pronounced shees.
[B] “Making a voyage” (Greenlandish) = secure a good cargo.
[C] Young bears are now regularly trained by the Eskimos for heavy sleigh work.
[D] The sea always looks black among or near the ice.—G. S.
[E] Raxed = stretched.
[F] Bield = shelter.
[G] Old-fashioned.

Typographical errors corrected by

the etext transcriber:
first disovered=> first discovered {pg
162}
shout of the b’s’n’s=> shout of the
bo’s’n’s {pg 139}
*** END OF THE PROJECT GUTENBERG EBOOK IN THE GREAT
WHITE LAND ***

Updated editions will replace the previous one—the old editions will
be renamed.

Creating the works from print editions not protected by U.S.

copyright law means that no one owns a United States copyright in
these works, so the Foundation (and you!) can copy and distribute it
in the United States without permission and without paying copyright
royalties. Special rules, set forth in the General Terms of Use part of
this license, apply to copying and distributing Project Gutenberg™
electronic works to protect the PROJECT GUTENBERG™ concept
and trademark. Project Gutenberg is a registered trademark, and
may not be used if you charge for an eBook, except by following the
terms of the trademark license, including paying royalties for use of
the Project Gutenberg trademark. If you do not charge anything for
copies of this eBook, complying with the trademark license is very
easy. You may use this eBook for nearly any purpose such as
creation of derivative works, reports, performances and research.
Project Gutenberg eBooks may be modified and printed and given
away—you may do practically ANYTHING in the United States with
eBooks not protected by U.S. copyright law. Redistribution is subject
to the trademark license, especially commercial redistribution.

START: FULL LICENSE

THE FULL PROJECT GUTENBERG LICENSE
 PLEASE READ THIS BEFORE YOU DISTRIBUTE OR USE THIS WORK

To protect the Project Gutenberg™ mission of promoting the free

distribution of electronic works, by using or distributing this work (or
any other work associated in any way with the phrase “Project
Gutenberg”), you agree to comply with all the terms of the Full
Project Gutenberg™ License available with this file or online at
www.gutenberg.org/license.

Section 1. General Terms of Use and

Redistributing Project Gutenberg™
electronic works
1.A. By reading or using any part of this Project Gutenberg™
electronic work, you indicate that you have read, understand, agree
to and accept all the terms of this license and intellectual property
(trademark/copyright) agreement. If you do not agree to abide by all
the terms of this agreement, you must cease using and return or
destroy all copies of Project Gutenberg™ electronic works in your
possession. If you paid a fee for obtaining a copy of or access to a
Project Gutenberg™ electronic work and you do not agree to be
bound by the terms of this agreement, you may obtain a refund from
the person or entity to whom you paid the fee as set forth in
paragraph 1.E.8.

1.B. “Project Gutenberg” is a registered trademark. It may only be

used on or associated in any way with an electronic work by people
who agree to be bound by the terms of this agreement. There are a
few things that you can do with most Project Gutenberg™ electronic
works even without complying with the full terms of this agreement.
See paragraph 1.C below. There are a lot of things you can do with
Project Gutenberg™ electronic works if you follow the terms of this
agreement and help preserve free future access to Project
Gutenberg™ electronic works. See paragraph 1.E below.
1.C. The Project Gutenberg Literary Archive Foundation (“the
Foundation” or PGLAF), owns a compilation copyright in the
collection of Project Gutenberg™ electronic works. Nearly all the
individual works in the collection are in the public domain in the
United States. If an individual work is unprotected by copyright law in
the United States and you are located in the United States, we do
not claim a right to prevent you from copying, distributing,
performing, displaying or creating derivative works based on the
work as long as all references to Project Gutenberg are removed. Of
course, we hope that you will support the Project Gutenberg™
mission of promoting free access to electronic works by freely
sharing Project Gutenberg™ works in compliance with the terms of
this agreement for keeping the Project Gutenberg™ name
associated with the work. You can easily comply with the terms of
this agreement by keeping this work in the same format with its
attached full Project Gutenberg™ License when you share it without
charge with others.

1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside the
United States, check the laws of your country in addition to the terms
of this agreement before downloading, copying, displaying,
performing, distributing or creating derivative works based on this
work or any other Project Gutenberg™ work. The Foundation makes
no representations concerning the copyright status of any work in
any country other than the United States.

1.E. Unless you have removed all references to Project Gutenberg:

1.E.1. The following sentence, with active links to, or other

immediate access to, the full Project Gutenberg™ License must
appear prominently whenever any copy of a Project Gutenberg™
work (any work on which the phrase “Project Gutenberg” appears, or
with which the phrase “Project Gutenberg” is associated) is
accessed, displayed, performed, viewed, copied or distributed:
 This eBook is for the use of anyone anywhere in the United
States and most other parts of the world at no cost and with
almost no restrictions whatsoever. You may copy it, give it away
or re-use it under the terms of the Project Gutenberg License
included with this eBook or online at www.gutenberg.org. If you
are not located in the United States, you will have to check the
laws of the country where you are located before using this
eBook.

1.E.2. If an individual Project Gutenberg™ electronic work is derived

from texts not protected by U.S. copyright law (does not contain a
notice indicating that it is posted with permission of the copyright
holder), the work can be copied and distributed to anyone in the
United States without paying any fees or charges. If you are
redistributing or providing access to a work with the phrase “Project
Gutenberg” associated with or appearing on the work, you must
comply either with the requirements of paragraphs 1.E.1 through
1.E.7 or obtain permission for the use of the work and the Project
Gutenberg™ trademark as set forth in paragraphs 1.E.8 or 1.E.9.

1.E.3. If an individual Project Gutenberg™ electronic work is posted

with the permission of the copyright holder, your use and distribution
must comply with both paragraphs 1.E.1 through 1.E.7 and any
additional terms imposed by the copyright holder. Additional terms
will be linked to the Project Gutenberg™ License for all works posted
with the permission of the copyright holder found at the beginning of
this work.

1.E.4. Do not unlink or detach or remove the full Project

Gutenberg™ License terms from this work, or any files containing a
part of this work or any other work associated with Project
Gutenberg™.

1.E.5. Do not copy, display, perform, distribute or redistribute this

electronic work, or any part of this electronic work, without
prominently displaying the sentence set forth in paragraph 1.E.1 with
active links or immediate access to the full terms of the Project
Gutenberg™ License.
1.E.6. You may convert to and distribute this work in any binary,
compressed, marked up, nonproprietary or proprietary form,
including any word processing or hypertext form. However, if you
provide access to or distribute copies of a Project Gutenberg™ work
in a format other than “Plain Vanilla ASCII” or other format used in
the official version posted on the official Project Gutenberg™ website
(www.gutenberg.org), you must, at no additional cost, fee or expense
to the user, provide a copy, a means of exporting a copy, or a means
of obtaining a copy upon request, of the work in its original “Plain
Vanilla ASCII” or other form. Any alternate format must include the
full Project Gutenberg™ License as specified in paragraph 1.E.1.

1.E.7. Do not charge a fee for access to, viewing, displaying,

performing, copying or distributing any Project Gutenberg™ works
unless you comply with paragraph 1.E.8 or 1.E.9.

1.E.8. You may charge a reasonable fee for copies of or providing

access to or distributing Project Gutenberg™ electronic works
provided that:

• You pay a royalty fee of 20% of the gross profits you derive from
the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”

• You provide a full refund of any money paid by a user who

notifies you in writing (or by e-mail) within 30 days of receipt that
s/he does not agree to the terms of the full Project Gutenberg™
License. You must require such a user to return or destroy all
 copies of the works possessed in a physical medium and
discontinue all use of and all access to other copies of Project
Gutenberg™ works.

• You provide, in accordance with paragraph 1.F.3, a full refund of

any money paid for a work or a replacement copy, if a defect in
the electronic work is discovered and reported to you within 90
days of receipt of the work.

• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.

1.E.9. If you wish to charge a fee or distribute a Project Gutenberg™

electronic work or group of works on different terms than are set
forth in this agreement, you must obtain permission in writing from
the Project Gutenberg Literary Archive Foundation, the manager of
the Project Gutenberg™ trademark. Contact the Foundation as set
forth in Section 3 below.

1.F.

1.F.1. Project Gutenberg volunteers and employees expend

considerable effort to identify, do copyright research on, transcribe
and proofread works not protected by U.S. copyright law in creating
the Project Gutenberg™ collection. Despite these efforts, Project
Gutenberg™ electronic works, and the medium on which they may
be stored, may contain “Defects,” such as, but not limited to,
incomplete, inaccurate or corrupt data, transcription errors, a
copyright or other intellectual property infringement, a defective or
damaged disk or other medium, a computer virus, or computer
codes that damage or cannot be read by your equipment.

1.F.2. LIMITED WARRANTY, DISCLAIMER OF DAMAGES - Except

for the “Right of Replacement or Refund” described in paragraph
1.F.3, the Project Gutenberg Literary Archive Foundation, the owner
of the Project Gutenberg™ trademark, and any other party
distributing a Project Gutenberg™ electronic work under this
agreement, disclaim all liability to you for damages, costs and
expenses, including legal fees. YOU AGREE THAT YOU HAVE NO
REMEDIES FOR NEGLIGENCE, STRICT LIABILITY, BREACH OF
WARRANTY OR BREACH OF CONTRACT EXCEPT THOSE
PROVIDED IN PARAGRAPH 1.F.3. YOU AGREE THAT THE
FOUNDATION, THE TRADEMARK OWNER, AND ANY
DISTRIBUTOR UNDER THIS AGREEMENT WILL NOT BE LIABLE
TO YOU FOR ACTUAL, DIRECT, INDIRECT, CONSEQUENTIAL,
PUNITIVE OR INCIDENTAL DAMAGES EVEN IF YOU GIVE
NOTICE OF THE POSSIBILITY OF SUCH DAMAGE.

1.F.3. LIMITED RIGHT OF REPLACEMENT OR REFUND - If you

discover a defect in this electronic work within 90 days of receiving it,
you can receive a refund of the money (if any) you paid for it by
sending a written explanation to the person you received the work
from. If you received the work on a physical medium, you must
return the medium with your written explanation. The person or entity
that provided you with the defective work may elect to provide a
replacement copy in lieu of a refund. If you received the work
electronically, the person or entity providing it to you may choose to
give you a second opportunity to receive the work electronically in
lieu of a refund. If the second copy is also defective, you may
demand a refund in writing without further opportunities to fix the
problem.

1.F.4. Except for the limited right of replacement or refund set forth in
paragraph 1.F.3, this work is provided to you ‘AS-IS’, WITH NO
OTHER WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PURPOSE.

1.F.5. Some states do not allow disclaimers of certain implied

warranties or the exclusion or limitation of certain types of damages.
If any disclaimer or limitation set forth in this agreement violates the
law of the state applicable to this agreement, the agreement shall be
interpreted to make the maximum disclaimer or limitation permitted
by the applicable state law. The invalidity or unenforceability of any
provision of this agreement shall not void the remaining provisions.
1.F.6. INDEMNITY - You agree to indemnify and hold the
Foundation, the trademark owner, any agent or employee of the
Foundation, anyone providing copies of Project Gutenberg™
electronic works in accordance with this agreement, and any
volunteers associated with the production, promotion and distribution
of Project Gutenberg™ electronic works, harmless from all liability,
costs and expenses, including legal fees, that arise directly or
indirectly from any of the following which you do or cause to occur:
(a) distribution of this or any Project Gutenberg™ work, (b)
alteration, modification, or additions or deletions to any Project
Gutenberg™ work, and (c) any Defect you cause.

Section 2. Information about the Mission of

Project Gutenberg™
Project Gutenberg™ is synonymous with the free distribution of
electronic works in formats readable by the widest variety of
computers including obsolete, old, middle-aged and new computers.
It exists because of the efforts of hundreds of volunteers and
donations from people in all walks of life.

Volunteers and financial support to provide volunteers with the

assistance they need are critical to reaching Project Gutenberg™’s
goals and ensuring that the Project Gutenberg™ collection will
remain freely available for generations to come. In 2001, the Project
Gutenberg Literary Archive Foundation was created to provide a
secure and permanent future for Project Gutenberg™ and future
generations. To learn more about the Project Gutenberg Literary
Archive Foundation and how your efforts and donations can help,
see Sections 3 and 4 and the Foundation information page at
www.gutenberg.org.

Section 3. Information about the Project

Gutenberg Literary Archive Foundation
The Project Gutenberg Literary Archive Foundation is a non-profit
501(c)(3) educational corporation organized under the laws of the
state of Mississippi and granted tax exempt status by the Internal
Revenue Service. The Foundation’s EIN or federal tax identification
number is 64-6221541. Contributions to the Project Gutenberg
Literary Archive Foundation are tax deductible to the full extent
permitted by U.S. federal laws and your state’s laws.

The Foundation’s business office is located at 809 North 1500 West,

Salt Lake City, UT 84116, (801) 596-1887. Email contact links and up
to date contact information can be found at the Foundation’s website
and official page at www.gutenberg.org/contact

Section 4. Information about Donations to

the Project Gutenberg Literary Archive
Foundation
Project Gutenberg™ depends upon and cannot survive without
widespread public support and donations to carry out its mission of
increasing the number of public domain and licensed works that can
be freely distributed in machine-readable form accessible by the
widest array of equipment including outdated equipment. Many small
donations ($1 to $5,000) are particularly important to maintaining tax
exempt status with the IRS.

The Foundation is committed to complying with the laws regulating

charities and charitable donations in all 50 states of the United
States. Compliance requirements are not uniform and it takes a
considerable effort, much paperwork and many fees to meet and
keep up with these requirements. We do not solicit donations in
locations where we have not received written confirmation of
compliance. To SEND DONATIONS or determine the status of
compliance for any particular state visit www.gutenberg.org/donate.

While we cannot and do not solicit contributions from states where

we have not met the solicitation requirements, we know of no
prohibition against accepting unsolicited donations from donors in
such states who approach us with offers to donate.

International donations are gratefully accepted, but we cannot make

any statements concerning tax treatment of donations received from
outside the United States. U.S. laws alone swamp our small staff.

Please check the Project Gutenberg web pages for current donation
methods and addresses. Donations are accepted in a number of
other ways including checks, online payments and credit card
donations. To donate, please visit: www.gutenberg.org/donate.

Section 5. General Information About Project

Gutenberg™ electronic works
Professor Michael S. Hart was the originator of the Project
Gutenberg™ concept of a library of electronic works that could be
freely shared with anyone. For forty years, he produced and
distributed Project Gutenberg™ eBooks with only a loose network of
volunteer support.

Project Gutenberg™ eBooks are often created from several printed

editions, all of which are confirmed as not protected by copyright in
the U.S. unless a copyright notice is included. Thus, we do not
necessarily keep eBooks in compliance with any particular paper
edition.

Most people start at our website which has the main PG search
facility: www.gutenberg.org.

This website includes information about Project Gutenberg™,

including how to make donations to the Project Gutenberg Literary
Archive Foundation, how to help produce our new eBooks, and how
to subscribe to our email newsletter to hear about new eBooks.