This book provides a comprehensive overview of Swedish phonology. There are two chapters on respectively vowels and consonants, describing the alternations and distributions of allophones and providing new, motivated phonemic charts. The rules and processes that are primarily segmental in nature are collected in one chapter, which includes both fully productive processes, and partly petrified ones. Among the most interesting segmental processes are the sandhi process of retroflexion, where interaction of several conditioning factors (quantity, morphology, phonation) accounts for the distribution, and the rule of d-continuization whereby a /d/ is realized as [r] in words that lack phonological stress. Most previous interest in Swedish phonology concerns the prosodic level and several chapters are devoted to an account for the interaction of stress, tone, prosodic structure and morphological structure. The minimal prosodic word is shown to be the domain of syllabification, of culminativity by stress, and of the assignment of a lexical tone from suffix to primary stress. Unlike English, Swedish only admits one stress per minimal prosodic word. Moreover, Swedish stress information is largely specified in morphemes (as stressed, or as subcategorizing for a stress). Together with a phonological stress rule, these factors greatly constrain word formation in Swedish, as well as account for the rather spectacular process of nickname formation. The argument for this organization of things is extensively laid out in two chapters focussing on the stress and accent distribution, respectively. Another chapter lays out the quantity system of Swedish, where stressed syllables must be heavy, but by virtue of containing either a long vowel or a long consonant. There is a chapter on the orthographic representation and comments on orthography are also given in the chapters on vowels and consonants. The book aims at providing new analyses where motivated, but at the same time endeavours not to be too theory dependent in presentation.

We give an overview of the phonological properties and processes that define the categories of the prosodic hierarchy in Swedish: the prosodic word (ω), the prosodic phrase (φ) and the intonation phrase (ι). The separation of two types of tonal prominence, big accents versus small accents (previously called focal and word accent, e.g. Bruce 1977, 2007), is crucial for our analysis. The ω in Swedish needs to be structured on two levels, which we refer to as the minimal ω and the maximal ω, respectively. The minimal ω contains one stress, whereas the maximal ω contains one accent. We argue for a separate category φ that governs the distribution of big accents within clauses. The ι governs the distribution of clause-related edge phenomena like the initiality accent and right-edge boundary tones as well as the distribution of nuclear big accents.

SUMMARYThis article presents a hypothesis about the origin of tone accent in Swedish and Norwegian. The main idea is that the presence of Proto-Nordic secondary stress and its subsequent reduction is essential to the development of accent 2. Developing an earlier proposal (Riad 1988), it is argued that stress clash is critical in the phonologization of tonal information. Support for the major claims is provided by the correlation of Proto-Nordic secondary stress and accent 2 in the modern languages, the synchronic behaviour of accent 2 in Modern Standard Swedish, its manifestation in conservative dialects, and distributional differences between three Scandinavian varieties, which reflect different stages of development. Other, earlier theories of the origin of the accents are discussed in the appendix.RÉSUMÉCet article présente une hypothèse de l'origine des accents tonaux en suédois et en norvégien. L'idée principale est que la présence de l'accent dynamique secondaire en proto-nordique et sa réduction subséquente sont essentielles pour le développement de l'accent 2. Ici l'auteur développe la proposition de Riad (1988) selon laquelle le facteur qui cause la phonologisation d'information tonale est le 'stress clash' (le conflit de l'accent tonique). À l'appui des aspects principaux de l'hypothèse il y a la corrélation entre l'accent dynamique secondaire en proto-nordique et l'accent 2 (tonal) dans les langues modernes, la comportement synchronique d'accent 2 en suédois moderne, sa manifestation dans les dialectes conservateurs, et des différences distribu-tionelles entre trois dialectes Scandinaves, qui présentent les différents étapes du développement. On discute d'autres théories de l'origine des accents dans un appendice.ZUSAMMENFASSUNGIn diesem Aufsatz wird eine Hypothèse zur ÖEntstehung der schwedischen und norwegischen tonalen Akzente vorgelegt. Ihr Kernpunkt ist, daß sekun-dare Betonungen im Urnordischen und ihre spätere Reduktion eng mit der Ent-wicklung des Akzents 2 verbunden sind. Als auslösenden Faktor in der Pho-nologisierung tonaler Information wird hier ein sog. 'Stress Clash' (nach Riad 1988) angenommen. Stütze für die hauptsächlichen Aspekte dieser Hypothèse finden wir in der Korrelation zwischen urnordischer sekundärer Betonung und Akzent 2 in den modernen Sprachen, dem synchronischen Verhalten des Ak-zent 2 im modernen Standardschwedischen, ihrer Manifestation in konserva-tiven Dialekten und auch in den distributionellen Kontrasten in drei skandi-navischen Varietäten, denen verschiedene Stadien der Entwicklung des Tons entsprechen. Andere Theorien zum Ursprung der Akzente werden in einem Anhang diskutiert.

Swedish stress and tone accent exhibits an interesting mixture of properties. I argue that the stress system is arranged in a largely morphological fashion, with clear similarities to dominance systems of Japanese, Basque and Greek, where there is a distinction between accented and unaccented stems, and where prefixes and, in particular, suffixes influence stress/accent placement. A major difference is that none of the lexical specifications for stress in Swedish is pre-or post-accenting, but rather post-and pretonic. Thus, no stress is assigned by affixes, but affixes impose adjacency conditions on stress placement in stems, or else the structure is either inhibited, or becomes noticeably marked. Beside the morphological specifications of stress information, there is a phonological default stress assignment, similar to what we find in Greek. The phonological default of Swedish applies blindly when prosodic specification is lacking at the right edge of prosodic words. An accentual default occurs also in Basque, but it applies at a phrasal level rather than at the word level.Beside stress, Swedish also exhibits a lexical tone ('accent 2', 'grave'), which occurs only in primary stressed syllables, and which (in the analysis assumed here) is mostly assigned from posttonic suffixes to an immediately preceding primary stress. So-called 'accent 1' (acute) is lexically unmarked, but both tonal contours signal prominence in a similar fashion, that is, in a way that is independent of the lexical distinction as such.Stress and tonal accent both instantiate culminativity. Building on the theory of projecting words and phrases (Itô and Mester 2007), I argue that stress instantiates culminativity within the minimal prosodic word, and tonal accent instantiates culminativity in the maximal prosodic word.

We propose an analysis of Greek meter based purely on phonology and the idea that well-formedness in meter is largely gradient, rather than absolute. Our analysis is surface-true, constraint-based and nonderivational, in line with proposals like optimality theory (Prince and Smolensky 1993). The discussion centers on two properties of meter, rhythm (dactylic, anapestic, iambic ...) and line length (hexameter, pentameter, tetrameter ...). Golston (1996) Unmarked meters are expected to be binary (dimeter) and rhythmic (no clash or lapse). We analyze individual meters in terms of how they deviate from this unmarked state, where deviations (big and small) are encoded directly as constraint violations following