1. Overview and structure of the book

The Manual of Romance Phonetics and Phonology is an attempt at providing a state-of-the-art overview of the most important works and discussions related to Romance varieties. The volume, as the editors put it (p. 20), is addressed both to those already familiar with the subject and to those who wish to become familiar with key facts and debates.

The book provides an expansive and well-thought-out discussion of key topics and exemplifies, in my view, an open approach to phonetic and phonological debate, without important omissions or adherence to a particular theoretical program or formal framework. It consists of 28 chapters organised into two parts, with contributions by 42 researchers and nearly 1000 pages of content. It certainly was a grand enterprise on the part of both the authors and the editors, the end result being a fairly functional and useful handbook, and a comprehensive reference material for all interested scholars. The book is very well prepared in terms of production quality – the figures and examples are well aligned with the text, references and IPA symbols are used correctly, and I do not see any typographical errors. At the end, the reader will find an index, which will make it easier to find particular topics in the right chapters.

The structure of the Manual is not very intuitive at first sight. The editors decided to divide the volume into only two parts: ‘Key topics in Romance phonology’ and ‘Phonetics and phonology of Romance languages’ – titles which, in my opinion, require further explanation of the contents. The first part consists of chapters concerning phonetic and phonological analysis across the language family, from specific developments concerning weight and syllable structure (Chapters 1–2), through aspects of segmental phonology and prosodic issues (Chapters 3–5 and 7–9, respectively), to methodological questions (Chapters 10–11) and aspects of first and second language acquisition (Chapters 12–14), variation (Chapter 15) and loanword phonology (Chapter 16). Part II of the book, on the other hand, is a compilation of chapters describing the phonetics and phonology of selected Romance vernaculars (12 chapters in total). Not all existing varieties are discussed here, but, in addition to a number of ‘mainstream’ Romance languages such as Italian, French, Spanish or Portuguese (with each chapter necessarily including discussions of dialectal variation), minority languages are also described in separate chapters, e.g. Galician, Occitan and Rhaeto-Romance varieties. Finally, Part II also provides a thorough examination of Judeo-Spanish as a representative of a diaspora variety, and an overview of Romance-based creoles. Thus, the Manual is quite comprehensive in terms of including detailed discussions of segmental and suprasegmental issues within particular languages in the second part as well as cross-linguistic and cross-dialectal comparisons of selected topics presented in the first part.

While the structure of the book, especially its first part, seems a bit confusing at first, after reading all of the chapters it becomes clear that many of the discussed phenomena are intertwined and reappear in different forms in several contributions. This recurrence, however, is not of an annoying type but rather serves as an interesting reminder that the many Romance varieties share a common history and some of their change trajectories coincide while others diverge across space and time. Besides, the recurring themes come back under different guises and often touch upon related but different issues or are viewed from different perspectives. For instance, length and weight in Romance languages (Chapter 1) are strictly related to the differential evolution of stress patterns across the Romance vernaculars (Chapter 7) and vocalic processes, such as vowel reduction (Chapters 3 and 4). Consonantal processes may also accompany or coincide with vocalic changes, e.g. in vowel nasalisation, an important change trajectory in French and Portuguese, among others (Chapter 6). Vowel changes, such as reductions, deletions, diphthongisation, etc., in turn, affect syllable structure (Chapter 2), and so on. Thus, upon reading the subsequent chapters of the Manual, one starts to feel as if puzzles of a bigger story were being put together one by one. Linguists specialising in one or several Romance languages will also find it satisfying to see the amount of detail in the provided descriptions, supplemented by a body of additional knowledge coming from related subfields, such as second language research, sociophonetics, studies of bi- and multilingualism, and language contact. Given the long history of the linguistic study of Romance, taking just one aspect of any given variety or any vernacular as a whole will always give an impression of incompleteness. Supplementing the segmental and prosodic phonology core with additional contexts of phonological development is therefore an added value.

2. Major points of interest and issues to consider

Given the length of the Manual, the number of chapters, the enormous amount of data and the number of descriptions, generalisations and analyses therein, it is extremely difficult to do justice to all the contributors and provide an exhaustive review of the material. For this reason, I will try to raise a number of issues that drew my attention based on particular parts of the volume. I hope that doing this in a chronological order will help the reader get the gist of at least some points of interest of the many that can be found in the Manual.

The book begins with an introductory chapter providing an overview of the fields of phonetics and phonology. These pages were both unexpected and refreshing. The editors decided to give a historical sketch of philosophy and research which gave rise to what we now know as the two subdisciplines of linguistics evoked in the title of the book. I find this chapter very interesting, especially for those young linguists who are looking for a succinct summary of the field and wish to understand the intricacies of our past and present understanding of phonemes, phonological processes and the changing approaches to the study of language, the role of data and experimentation, as well as attitudes towards variation and change. Notwithstanding the value of this introduction, I feel that it could include a geographical overview of Romance varieties, perhaps with a map or a sketch that would familiarise readers with the particular tongues and their territorial distributions, especially since the first chapters run through a multiplicity of vernaculars and their particularities, without detailed explanation or introduction.

Shifting focus to the chapters devoted to the phonological core of Romance linguistics, i.e. segmental and suprasegmental issues, it is worth noting that these are mostly descriptive. They are all structured in a similar way, providing a historical background of such topics as the development of syllable structure, changes in syllable weight and vowel length and shifts in the properties of segments derived from Latin forms. Different paths of the different subgroups of Romance languages are presented in separate subsections. Also, the chapters as a whole do not espouse any particular theory or framework. Rather, a variety of different perspectives is presented, from autosegmental theory, Dependency Theory, Government Phonology and related frameworks, through Articulatory Phonology to Optimality Theory and theories of intonation (Dependency Incremental and Autosegmental-Metrical frameworks).

As for the presentation style, I particularly liked Chapter 1 by Pavel Iosad and Leo Wetzels, who provide an impressive review of the historical developments of weight and length in the different Romance vernaculars, pointing to three important questions related to the topic: weight sensitivity, the minimal word requirement and the three-syllable window. Their argumentation, based on theoretical and empirical studies involving truncation and treatment of nonce words, acronyms and loans by native speakers, is particularly convincing and leads to the conclusion that syllable weight is an important explanatory concept in Romance. The chapters on nasalisation and palatalisation also deserve special attention. In the latter case, the authors do not stop at providing a purely descriptive analysis. Instead, they confront the data with several theoretical accounts, ensuring useful discussion with the literature. The proposed Articulator Theory-based framework by Calabrese (Reference Calabrese2005) is an interesting solution that seems to unify the different patterns of variation across Romance (and beyond). The reader should carefully analyze the presented arguments, however. For instance, the No Tampering Condition (p. 182) is a bit confusing with respect to Articulatory Phonology vis-‘a-vis Chomsky's (Reference Chomsky, Freidin, Otero and Zubizarreta2008) definition of the term, although it does coincide with the latter conceptually. This might have been mentioned by the authors, especially when used as a major argument against Recasens's (Reference Recasens2011) account of palatalisation. Also, in their criticism of Rubach's (Reference Rubach2003) OT account, the authors oversimplify the OT formalism and then argue against the proposal by invoking the putative implausibility of fixed major place feature ranking. Nevertheless, there is ample evidence in phonology for the greater markedness of labials and dorsals compared to coronals (cf. Prince & Smolensky Reference Prince and Smolensky1993). This is in line with the observed facts in Romance, as noted by the authors (no labialisation in palatalisation processes, p. 205). However, other places of articulation are not necessarily universally fixed, as suggested and criticised in point (18) of the chapter. Rather, OT would assume language-specific rankings in this case. All in all, the analysis of palatalisation presented in Chapter 5 attempts to link the formal mechanism responsible for the derivations to concrete information on the articulatory bases of the process as a two-step mechanism (raising and coronalisation). The readers must evaluate for themselves whether the complex featural changes and constraints on articulatory configurations included in Calabrese's model are explanatory and convincing enough compared to other proposals.

Chapters 10 and 11, focused on methodological issues, are a very useful addition to the Manual. It is a shame that this collection is so modest – I believe more review papers of this type would be much appreciated, e.g. a piece on Laboratory Phonology, fieldwork or frequency-based approaches. The chapter on corpus phonology is a very accessible introduction to statistical data analysis, ideal for beginners in the field. The contribution on sociophonetics contains both an overview of the methods and an exhaustive summary of those studies in Romance linguistics that focus on the social markers and determinants of particular processes. I definitely recommend both chapters to those interested in the mentioned approaches.

The editors of the Manual found it particularly important to pay quite a lot of attention to language acquisition in relation to Romance varieties. The volume includes not one but three chapters on various aspects of the topic (‘First language acquisition of Romance phonology’, ‘Bilingual phonological acquisition’ and ‘Second and third language acquisition of Romance phonology’) and two additional chapters whose themes are at least indirectly related to it (‘The phonology of Romance contact varieties’ and ‘Loanword phonology in Romance’). Each of these chapters gives a comprehensive overview of the field and key theories related to the particular topic, as well as a summary of the most important empirical studies. Since these are all separate chapters, a general discussion of acquisition in its many aspects is a bit lacking. Several aspects of language acquisition, be it L1 or subsequent languages, appear as important factors on several pages, but a link between them would be desirable. For instance, the perception of contrasts in early childhood, the amount of input received from the environment, as well as its quality play a role both in first and in second language acquisition. Johanna Stahnke (Chapter 12) provides a very thorough review of the patterns of first language acquisition in the context of Romance phonologies. Her conclusion is that segments tend to be acquired in accordance with Jakobson's (Reference Jakobson1941) universals; that is, plosives and nasals are acquired first. Also, voicing contrasts are acquired in a stepwise manner, with the use of a short-lag range of VOTs when producing target prevoiced stops. Additionally, CV syllables are preferred and complex margins are developed later on. The research on rhythmic and stress patterns, on the other hand, reveals that language-specific trajectories should be sought rather than cross-linguistic tendencies, even within the interrelated Romance family. Despite this discrepancy between universality and language-specificity in segmental vs prosodic aspects of phonology, however, the timing of acquisition uncovers a uniform trend: prosody is acquired first, and target-like segmental production follows.

One of the conclusions drawn by Stahnke strikes me as a bit counterintuitive. On page 396, it is stated that ‘Under the assumption that phonological acquisition is language-specific if “economical”, we may conclude that the acquisition of prosody is less “costly” than the acquisition of segments’. But why would the acquisition of universal properties be more costly than the acquisition of properties pertaining to a particular language? Universals express less marked and simpler structures and the assumption is that they are a part of an innate language instinct, faculty or predisposition, however we may understand the notion. I would therefore expect the tendency to produce stops or nasals and CV syllables first to be relatively easier and more straightforward. Nevertheless, there is plentiful evidence of the incredible ease with which infants perceive, calculate and learn patterns from the very first months of life. Perhaps this makes the language-specific process more ‘economical’ as a result. Some of this ability was mentioned by Stahnke, but I find important literature lacking. For instance, there is an ongoing debate on how children acquire the phonological properties of language and chunk speech into meaningful units, with two competing hypotheses: stress and rhythm-based inference, including the so-called trochaic bias mentioned in the chapter (Jusczyk et al. Reference Jusczyk, Houston and Newsome1999; Thiessen & Saffran Reference Thiessen and Saffran2007) or phonotactic statistical learning, according to which infant brains calculate the probabilities of certain sound combinations within and between words (Saffran et al. Reference Saffran, Aslin and Newport1996, Reference Saffran, Senghas and Trueswell2001; Mattys & Jusczyk Reference Mattys and Jusczyk2001). The cooccurrence of phonotactic statistical learning patterns with the rhythm-based one points to the possibility that not all segmental acquisition trends are universal. They may be modulated by language- or input-specific patterns, as demonstrated by Conxita Lleó in Chapter 13 on bilingual acquisition, in which the acquisition of codas differed in Spanish–German bilinguals compared to Spanish monolinguals due to the higher frequency of codas in the input. The latter chapter provides a different perspective by showing mostly results of particular empirical studies. As such, it is a well-organised summary of the phonological development of segmental and prosodic structures in bilinguals, complementary to the review of changes provided in the preceding chapter by Stahnke. However, it is a bit of a shame that it did not explore bilingualism in diglossic communities, fully bilingual communities and monolingual communities from a comparative perspective (e.g. Switzerland, Northern Italy or Luxembourg would be good case studies). It would be interesting for senior and junior researchers alike to explore the differences in phonological development based on community type.

In Chapter 14 on second- and third-language acquisition, Nicholas Henriksen complements Chapters 12 and 13 with a very detailed discussion of the literature, as well as theoretical models proposed to address the topic, such as Flege's (Reference Flege and Strange1995) Speech Learning Model or Best et al.'s Perceptual Assimilation Model (Best Reference Best and Strange1995; Best & Tyler Reference Best, Tyler, Bohn and Munro2007). The chapter is impressive in the number of studies it discusses in the context of L2/L3 Romance languages. Perhaps the most interesting part concerns language transfer and the way each acquired language influences others (or not). According to Henriksen, while in the context of L1 and L2 we can talk of bidirectional influence in some cases, L3 is usually not influenced by L1 but by L2, i.e. a previously acquired foreign language. The interpretation of these findings is somewhat obscured by the fact that L2 and L3 are usually closely related languages in the reviewed studies. I highly recommend Henriksen's chapter, as it shows very interesting dependencies between the different Romance languages, places them in a broader context of the proposed theoretical models and points to important factors of language transfer, such as frequency of use, language dominance and proficiency. These important notions appear here and there in the three acquisition chapters, but what I find lacking is a broader contextualisation within the general acquisition theory framework. In the context of bi- and multilingualism, it is often emphasised that the very definition of the terms is difficult and should be as broad as possible, including L2 and L3 learners (Grosjean Reference Grosjean2010). Furthermore, it is argued that language acquisition is governed by the combination of age of acquisition, proficiency and cognitive control mechanisms (see Hernandez Reference Hernandez2013 for an overview). It also evolves in accordance with the child's brain development (cf. Piaget's Reference Piaget, Green, Ford and Flamer1971 model). Neuroscientific research shows that sensorimotor processing is responsible for the first stages of speech perception and production, and as the brain develops, different areas begin to be involved, especially the prefrontal cortex, whose maturation modulates learning mechanisms and cognitive functions related to working memory and attention switching. Most importantly, these changes are responsible for the fact that adults and children learn languages differently and that language acquisition is fundamentally different in non-simultaneous bilinguals. Crucially for the acquisition of phonology, our perception of non-native contrasts and ability to reproduce them is largely inhibited after the first years of childhood, together with the implicit learning of grammatical rules, as opposed to explicit grammar learning and vocabulary acquisition, which we can master at any age.

As for Part II of the Manual, it includes a series of thorough descriptions of particular languages. In this case, the repertoire of vernaculars is smaller than in the numerous mentions in the theoretical chapters from Part I, but the effect is impressive nonetheless. Each of these chapters constitutes a short but exhaustive description of the segmental and suprasegmental phonetics and phonology of a given language or languages. They are given by experts in the field, specialising each in a particular variety and having numerous publications on the subject, and constitute a great summary of the processes discussed in Part I. Worthy of particular attention are chapters that discuss phonological problems that are subject to current debate in the literature, such as the status of the schwa in French (Chapter 21 by Marie-Hélène Côté) or Martin Krämer's (Chapter 18) account of Italian, which discusses experimental and theoretical evidence concerning the productivity of velar palatalisation, the syntax–phonology interactions related to raddoppiamento sintattico and domain-final lengthening, as well as the controversial question of stress assignment. In the latter case, it is shown that the adoption of trimoraic layered feet as per Martínez-Paricio (Reference Martínez-Paricio2013) helps explain the discrepancies between bisyllabic and trisyllabic words with different stress patterns.

3. Overall evaluation

The Manual of Romance Phonetics and Phonology is a comprehensive compendium of key topics in Romance varieties, both from a diachronic and a synchronic perspective. It thoroughly investigates segmental and prosodic issues related to a vast range of Romance languages and dialects, providing a comparative perspective and an in-depth view of variation and change, and employing a wide array of formal approaches to segmental and suprasegmental processes.

As such, the volume will be of interest for linguists and students looking for a reference manual containing detailed information on any of the many Romance varieties that may be the subject of phonetic and phonological endeavour. It will also be very useful for those who look for thorough accounts of particular problems across Romance, e.g. the converging and diverging behaviours of vowels or consonants, or differences in syllable structure and stress patterns. Furthermore, those wishing to broaden their knowledge on a particular topic beyond their specialisation – e.g. phoneticians interested in second language acquisition, phonologists interested in sociophonetic and language contact aspects of the studied phenomena – will find the volume very enriching.

The hallmark of this book is the way the contents of the chapters, and the multiplicity of topics and perspectives, intertwine throughout the reading experience. Although most readers will not read the entire Manual, they would hardly feel that any part was unnecessarily repetitive or redundant despite the fact that many themes are analysed in some way in multiple chapters. Some will perhaps seek a bird's-eye view of the many intricacies of Romance linguistics in a more general and succinct form, but this would be an impossible task given the broad nature of the subject matter and the overwhelming literature.