Allison Wetterlin

Lexical specification of tone in North Germanic

onsson-Steiner Accent1isverymuchacceptedintheliteratureasthedefaulttonalmarkerinScandinavian languages. Consequently, stems and affixes are almost always specified for accent 2. Only rarely in some analyses is accent 1 specified for affixes, but never for stems. We believe that under these conditions, the resulting morphology/phonology interaction is rather complex, having to include special rules of accent marking, floating tones, deaccenting together with inexplicable exceptions. In our analysis of the tonal systems of Swedish and Norwegian,accent1isthelexicallyspecifiedaccentandaccent2ispostlexicallyassigned. Words and affixes may be lexically specified for accent 1, which inevitably dominates. Consequently, if a morphologically complex word includes a lexically specified affix or stem, the entire word will bear accent 1, giving us patterns of alternations like beskriva1, skriva2. This analysis enables us to account for all the facts almost exceptionlessly, with no special tonal rules, constraints or templates.

Sounds Definite-ly Clitic: Evidence from Scandinavian Tone

Historically, the {en} defi nite ending comes from a demonstrative, the meaning of which was «weakened» with concomitant destressing (Wessen 1970). The syntactic double marking is conspicuous, but the advent of the defi nite ending also had vital consequences for tonal alternations. Early Scandinavian predictably distinguished two pitch accents based on syllable structure – monosyllabic words with accent 1, polysyllabic words with accent 2. Scandinavian scholars maintain that when the defi nite became an enclitic and attached to the end of a word, minimal pairs with a tonal opposition appeared as seen in (2) (cf. Oftedal 1952).

Processing of Phonemic Consonant Length: Semantic and Fragment Priming Evidence from Bengali.

Six cross-modal lexical decision tasks with priming probed listeners’ processing of the geminate–singleton contrast in Bengali, where duration alone leads to phonemic contrast ([pata] ‘leaf’ vs. [pat:a] ‘whereabouts’), in order to investigate the phonological representation of consonantal duration in the lexicon. Four form-priming experiments (auditory fragment primes and visual targets) were designed to investigate listeners’ sensitivity to segments of conflicting duration. Each prime derived from a real word ([kʰɔm]/[gʰenː]) was matched with a mispronunciation of the opposite duration (*[kʰɔmː]/*[gʰen]) and both were used to prime the full words [kʰɔma] (‘forgiveness’) and [gʰenːa] (‘disgust’) respectively. Although all fragments led to priming, the results showed an asymmetric pattern. The fragments of words with singletons mispronounced as geminates led to equal priming, while those with geminates mispronounced as singletons showed a difference. The priming effect of the real-word geminate fragment was significantly greater than that of its corresponding nonword singleton fragment. In two subsequent semantic priming tasks with full-word primes a stronger asymmetry was found: nonword geminates (*[kʰɔmːa]) primed semantically related words ([marjona] ‘forgiveness’) but singleton nonword primes (*[gʰena]) did not show priming. This overall asymmetry in the tolerance of geminate nonwords in place of singleton words is attributed to a representational mismatch and points towards a moraic representation of duration. While geminates require a mora which cannot be derived from singleton input, the additional information in geminate nonwords does not create a similar mismatch.