This volume is a collection of papers which brings together international researchers in the fields of morphology, morphosyntax, Romance languages, historical linguistics, linguistic typology, and dialectology. As the title suggests, the topic of this book, devoted to morphology and m orphosyntax, is the description and analysis of pure morphological units – units whose forms depend neither on semantics nor on phonology – and the reasons for the existence of such structures, which rely at times on diachronic and etymological analyses, in addition to their synchronic dimension.

The book's chapters centre on a fact that has become crucial in theoretical discussions of morphology: the formal autonomy displayed by certain morphological units with respect to semantic properties, syntactic functions or phonological matches. Aronoff (Reference Aronoff1994:25) provides a definition for this autonomy, i.e. the so-called ‘morphomic level of morphology’, which establishes the independence of morphology from the intersection of syntax/phonology. Thus, morphological entities which cannot be associated with constant meaning or be explained by the activation of phonological constraints or correspondence with any particular syntactic function (see also Spencer, to appear) are called morphomes.

Apart from noting the existence of purely formal, or morphomic, stems, Aronoff (Reference Aronoff1994) identifies two types of morpheme organising principles: syncretisms in inflectional paradigms (Baerman, Brown & Corbett Reference Baerman, Brown and Corbett2005) and the root-and-pattern organisation of Semitic verbs. He illustrates the latter former with the ‘third stem’ in Latin, but work on other languages can also be cited, e.g. studies on French verb and adjective inflection paradigms in Bonami & Boyé (Reference Bonami and Boyé2003, Reference Bonami and Boyé2005), where morphomic distribution is displayed in so-called stem spaces.

In this volume, a large portion of which is dedicated to inflection, the concept of morphome is complemented by other key notions, addressed in nearly all chapters. These notions have been introduced in Maiden (Reference Maiden2004, Reference Maiden, Booij and van Marle2005) to describe some verbal paradigms in Romance languages: the PYTA morphome (Perfecto Y Tiempos Afinos), and the U-, L- and N-patterns. In the latter set, the letters identifying the pattern evoke the various forms which stem from the morphomic stem distribution in verb conjugation in Romance languages.

This volume includes 13 chapters, preceded by an introduction. In many of the chapters the question of autonomy in morphology is illustrated by phenomena taken from various Romance languages and dialects. Different aspects of the notion of morphome are discussed: What is the scope of this concept? How does it manifest in the morphological constructions presented in the languages described? Is Aronoff's definition appropriate for the facts under investigation or is it too restrictive and should be extended? Finally, is the concept of morphome independent of any specific morphological theory? In other words, what is the impact of this notion on linguistic thought? In the remainder of this review I outline the content of the chapters and evaluate the volume.

Stephen R. Anderson's chapter is a study of a dialect of Swiss Romanche, the Surmiran of Savognin, in which the author argues against the notion of morphome, showing that allomorphic variations of verbal stems in this dialect are well-motivated. Stem shape is partly predictable by the stress pattern, itself phonologically conditioned. This analysis, which is also applied to other dialects of Romanche, confronts earlier conclusions, presented by Maiden (Reference Maiden, Smith, Goldbach and Hinzelin2011), for whom these variations are morphomic (this chapter is a new step in a dialogue between the two researchers, which began in 2008).

In his article, Martin Maiden addresses the notion of morphologisation, which he discussed in earlier work (e.g. Maiden Reference Maiden2009). Morphologisation affects phenomena that characterise a word, historically, once they cease to be phonologically motivated. Morphology takes over when the phonological environmental trigger disappears. When the environment remains in place, the loss of the phonological role can still be empirically demonstrated through the observation of similar environments and the identification of variations in the absence of the triggering environment. Based on these facts, Maiden re-evaluates the impact of phonology, which he nevertheless acknowledges plays a non-zero role in the verbal systems of Italian and Romanian.

Martina Da Tos is interested in Italian verbs of the inflectional class of finire ‘to end’, which present an alternation between two stems, the former having a shape X and the latter a shape Xisc. As this alternation cannot be explained as a stem morphomic variation in the N-pattern, the author considers the history of these verbs. Adopting a morpheme-based approach to morphology, she shows that the sequence sc, while being morphomic, leads not to a root variation but to a variation of inflectional marking.

Claire Meul employs the statistical tools of corpus linguistics to examine variation in a Ladin dialect spoken in the central Dolomites, where the stem of verbs of the first conjugation pattern is characterised by a variation employing the infix -id(i)-. This study is remarkable, since it combines field work (interviews with informants) with synchronic, etymological and historical analysis of the phenomenon, and integrates a sociolinguistic dimension.

Louise Esher studies the synthetic future and conditional forms in Occitan in order to explain the gap between formal similarities in the realisation of these tense forms and the differences in their interpretations. The author compares the results obtained in Occitan with data from other Romance languages, particularly the relation observed between form and semantic function, and concludes that this data instantiate a weak morphomic pattern, which interacts with the semantic and phonological levels. This leads her to propose a non-canonical definition of the morphome, but at the same time to emphasise the theoretical difficulty in combining morphome- and morpheme-based theories.

Nigel Vincent follows a similar direction. He too examines the differences between the form and interpretation of the conditional, particularly in Spanish, French and Italian. His aim is theoretical, based on the default principle of language compositionality. The author shows that the definition of morphomes must be extended, and concludes that that the use of the notion of morphomes and, consequently, the use of the notion of compositionality necessitate the adoption of an appropriate linguistic theory.

Michele Loporcaro describes the verbal inflection of Sardinian, in particular the paradigmatic organisation of irregularities in the Logudorese dialect, adopting the model of the Overall Distribution Schema (Pirrelli & Battista Reference Pirrelli and Battista2000). The author examines the structures of this dialect synchronically and diachronically, considering the extent to which L- and N-patterns are used. He concludes with a shallow analysis of other Sardinian dialects which are consistent with the results obtained in Logudorese.

Mark Aronoff investigates the nature of roots in Semitic languages as linguistic entities, contrasting two theoretical approaches, Distributed Morphology (Halle & Marantz Reference Halle, Marantz, Hale and Keyser1993) and Word-based Morphology (e.g. Blevins Reference Blevins2006). Following a historical overview of grammatical traditions in Arabic and Hebrew, the author shows through experimental data that these roots, which have no constant semantic content, are morphomes, showing similarities with stems in English and Romance languages.

Steven Kaye examines morphomic verbal stems in Talysdhi, a Northwestern Iranian language, from a diachronic and a synchronic point of view. As in other cases examined in this volume, the morphomic analysis depends on theoretical choices, revealing a much more complex situation than was previously described for this language.

Chiara Cappellaro focuses on a typical case of morphome, the notion of overabundance – the presence of multiple competing forms fulfilling the same cell of the verbal paradigm. The author applies this notion to third person pronouns, from a diachronic and an acquisitional perspective.

Paul O'Neill formalises the morphome as a purely morphological feature and compares it with morphosemantic and morphosyntactic features, which he renames morphemic features. Based on Latin, Spanish and English, he shows that there is no fundamental qualitative difference between morphomic and morphemic features, as both can manifest a discontinuous relation between form and meaning.

In the same vein, John Charles Smith establishes a typology of morphomes. He distinguishes two types, overt morphomes and covert morphomes. The first present regular patterns in paradigms independent from extramorphological factors while the latter correlate in some way with an extramorphological function. A review of the morphomic properties of L-, U-, and N- patterns, tense-aspect-mood (TAM), person, suppletion and defectiveness, leads the author to a hierarchical classification of morphomes based on the coherence of the structural types identified.

In the final chapter, Silvio Cruschina addresses morphological autonomy through inflectional periphrases. These constructions are peculiar in nature; given their grammatical properties, they cannot be entirely analysed syntactically, nor are they suited for a fully inflectional integration approach, given the existence of compositionality phenomena. Double-inflected verbal constructions in Sicilian, whose range of realisations is presented, serve to illustrate that it is the degree of grammaticalisation of these periphrases that enables the determination of their potential morphomic nature.

As is evident from the brief summaries above, the authors pursue varied objectives. Some focus on the analysis of a specific language in which the inflection system presents morphomic properties (or, alternatively, properties that counter this notion); others compare several languages and/or several historical stages of the same language, in order to study the mechanism leading to the autonomisation of morphology; finally, some authors propose a typology of morphomes or a new definition of this notion.

It is a pity that the volume, with a few exceptions, is largely devoted to inflection and does not elaborate on the questions of derivation, thus rendering it to a certain extent a continuation of the project in the Maiden et al. (Reference Maiden and Maiden2011) collection rather than as an entirely new contribution in its own right. Nevertheless, in its diversity, this book contributes to the vital reflection necessary for the understanding of current debates in morphology, addressing the nature of morphological units, the status of the processes that manipulate them and the relations between morphology and related disciplines.

At the present time, indeed, one of the most debated topics in morphology concerns its interfaces with syntax, phonology and semantics. Therefore, current discussions in morphology address questions such as the interaction between morphological rules and syntactic categories, morphosyntactic properties, segmental and prosodic features, and aspects of the denotation of expressions these rules may be sensitive to. Thus, showing that the autonomy of morphology is always offset by a set of limitations, as demonstrated in this volume, provides a new perspective on the place of morphology among the linguistic disciplines.

Moreover, the notion of morphome becomes measurable through formal and theoretical tools intended to assess the real extent of the phenomena and the linguistic frameworks under consideration, at least in the languages examined. In other words, this book offers new approaches to the understanding of these interactional considerations.

The volume thus constitutes a valuable contribution, highlighting the complexity behind the interactions among linguistic domains, and bringing new insights to account for these interactions.