Against contour tone units S Duanmu Linguistic inquiry 25 4 , , Metrical and tonal phonology of compounds in two Chinese dialects S Duanmu Language, , Syllabic weight and syllabic duration: A correlation between phonology and phonetics S Duanmu Phonology 11 1 , , Stress and the development of disyllabic words in Chinese S Duanmu Diachronica 16 1 , , Chinese Mandarin : Phonology S Duanmu. Start on.
Show related SlideShares at end. WordPress Shortcode. Published in: Technology , Education. Full Name Comment goes here. Are you sure you want to Yes No. Aurora Brennan Gym guru exposes fat burning secret! I'm going to reveal to you the secret method that allows you to get the equivalent exercise of 45 minutes at the gym, in just a few minutes per day! Thank you, thank you so much! Faiza Aleem , Huma Kayani. Rajeswari AK. Show More.
No Downloads. Views Total views. Actions Shares. Embeds 0 No embeds. No notes for slide. Syllabe 1. Nuclear elements are the vowels or syllabic segments; marginal elements are the consonants or non-syllabic segments. Thus we can define syllable as… A syllable is a speech unit that can be divided into two parts on set and rhyme within the rhyme there are nucleus a vowel and the coda an ending consonant 8. Thus nuclear elements , or syllabic segments can be described as intrinsically more sonorous than marginal, or non-syllabic elements.
Finally, antepenultimate stress words always carry orthographic accent.
Specifically, stress is morpho-syntactically conditioned in Spanish verbs. To my knowledge, there is not much research on the phonology of brand names, but the general observation is that good, successful brand names are generally easy to pronounce and conform to markedness constraints and phonotactics of a language see e. Therefore, if participants were interpreting the stimuli as product names, this would have made them more conservative in their judgments across the board on the antepenultimate stress conditions that are unattested in the lexicon. Manual inspection of the randomization orders does not show any clustering of items of a particular condition in any subject at any point during the task.
Correlations between the randomization orders and the script order in which the items are grouped by condition are not greater than 0. The results on the experimental task show a slight tendency reflecting this claim: the mean for antepenultimate stressed nonce words with voiceless final onsets was 2. I leave the exploration of this pattern for further experimental research. The grapheme-to-phoneme rules that the script implemented were designed together with the author.
A separate set of analyses with a lexicon in which the trill was coded as a geminate tap was performed. The results were very similar to the model in which both segments were codified as separate phonemes, and the correlations with the experimental data were also not significant.
The results and the correlations with the experimental results were not significantly different from the correlations obtained when the whole lexicon was used. For that reason, I report the original results, which correspond to comparisons to the whole lexicon. On this basis, all the maximum entropy models presented in this paper have been implemented both with and without token frequency weighting. I did not find a significant difference between the two kinds of models, so the results reported in here correspond to the initial models, which include token frequency weighting. This indicates that limiting the number of constraints to is sufficient to pick up both high and low weighted phonotactic constraints in Spanish.
All runs within each model were highly correlated, so in all cases the penalty scores were averaged over 10 runs. Once the predicted scores are ranked, there is no possible effect on the ranking by dividing the predicted scores by a Z value. Under this account, however, this can be explained by claiming that the constraint would have a high penalization score, but it could still be violated by some items.
However, most constraints will only refer to a 3-segment window, given that constraints that refer to dependencies between the nucleus of a heavy syllable and the following onset are unlikely.
In any case, running this specific model with the maximum constraint size set a 3 predicts ratings that have the same correlations both with respect to the whole set of conditions in the experimental data and with respect to the antepenultimate stress conditions. This shows that the set of constraints obtained under that window size produces the same results that when the constraint search space is increased. I would also like to thank Ryan Bennett for his continuous support and for the thorough discussion along all stages of this project.
Many thanks to Kevin Tang for comments on earlier drafts of this work and for his assistance with the grapheme-to-phoneme conversion for the lexicon in the analogical models. All errors and omissions remain of course my own. Albright, Adam. Feature-based generalisation as a source of gradient acceptability. Phonology 26 1.
Modeling analogy as probabilistic grammar. Oxford: Oxford University Press. Rules vs. Cognition 90 2. Alderson, Phil. Absence of evidence is not evidence of absence: We need to report uncertain results and do it clearly. British Medical Journal Absence of evidence is not evidence of absence. Alvord, Scott. The psychological unreality of quantity sensitivity in Spanish: Experimental evidence. Southwestern Journal of Linguistics 22 2. Aske, Jon. Disembodied rules versus patterns in the lexicon: Testing the psychological reality of Spanish stress rules.
Determinants of wordlikeness: Phonotactics or lexical neighborhoods?
In This Article
Journal of Memory and Language 44 4. Abstractness and motivation in phonological theory. Studies in Hispanic and Lusophone Linguistics 2 1. Exceptionality in Spanish stress. Catalan Journal of Linguistics A fresh look at quantity sensitivity in Spanish. Linguistics 40 2. Barker, Muhammad Abd-al-Rahman. Klamath grammar University of California Publications in Linguistics Fitting linear mixed-effects models using lme4.
San Duanmu - Google 學術搜尋引用文獻
Journal of Statistical Software 67 1. The Surfeit of the stimulus: Analytic biases filter lexical statistics in Turkish laryngeal alternations. Language 87 1. Beckman, Jill. Positional faithfulness, positional neutralisation and Shona vowel harmony.
Chapter Six. Geminates: Heavy Or Long?
Phonology 14 1. Segmental duration and rhythm in Spanish. Journal of Phonetics Bradley, Travis. A typology of rhotic duration contrast and neutralization. Probus 18 1. Brown, Charles. A grammar of the Telugu language. New Delhi: Asian Educational Services. Proceedings from the annual meeting of the Chicago Linguistic Society Explaining sonority projection effects. Phonology 28 2. BuscaPalabras: A program for deriving orthographic and phonological neighborhood statistics and other psycholinguistic indices in Spanish.
Behavior Research Methods 37 4. EsPal: One stop shopping for Spanish word properties. Behavioral Research Methods Eddington, David. Spanish stress assignment within the Analogical Model of Language.
Language 76 1. A computational approach to resolving certain issues in Spanish stress placement.