Japanese Dialects

Hattori, Shirô

doi:10.1515/9783111418797-017

Cited by 10 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…on Greenhill, Blust, and Gray 2008), and Hattori (1973) on Japanese dialect (data in electronic form supplemented in List 2014), an additional data set of 14 Chinese varieties originally published by Hóu (2004) was specifically modified and manually aligned for this study. While the two former data sets are classical wordlists that are further coded for cognacy and alignments, 21 the Chinese data is based on a collection of 623 morphemes (reflected by a Chinese character each) whose pronunciation across the 14 dialects used in our sample was elicited by field workers.…”

Section: General Characteristicsmentioning

confidence: 99%

Automatic Inference of Sound Correspondence Patterns across Multiple Languages

List

2019

Computational Linguistics

View full text Add to dashboard Cite

Sound correspondence patterns play a crucial role for linguistic reconstruction. Linguists use them to prove language relationship, to reconstruct proto-forms, and for classical phylogenetic reconstruction based on shared innovations. Cognate words that fail to conform with expected patterns can further point to various kinds of exceptions in sound change, such as analogy or assimilation of frequent words. Here I present an automatic method for the inference of sound correspondence patterns across multiple languages based on a network approach. The core idea is to represent all columns in aligned cognate sets as nodes in a network with edges representing the degree of compatibility between the nodes. The task of inferring all compatible correspondence sets can then be handled as the well-known minimum clique cover problem in graph theory, which essentially seeks to split the graph into the smallest number of cliques in which each node is represented by exactly one clique. The resulting partitions represent all correspondence patterns that can be inferred for a given data set. By excluding those patterns that occur in only a few cognate sets, the core of regularly recurring sound correspondences can be inferred. Based on this idea, the article presents a method for automatic correspondence pattern recognition, which is implemented as part of a Python library which supplements the article. To illustrate the usefulness of the method, I present how the inferred patterns can be used to predict words that have not been observed before.

show abstract

Section: General Characteristicsmentioning

confidence: 99%

Automatic Inference of Sound Correspondence Patterns across Multiple Languages

List

2019

Computational Linguistics

View full text Add to dashboard Cite

show abstract

“…Although p-initial words in modern Japanese are restricted to mimetics and foreign loan words, initial p is found in several dialects of Ryukyuan (Miyako, Yaeyama, Northern Okinawan) corresponding to initial h in standard Japanese and the Shuri (Southern) dialect of Okinawan (Hattori 1973;Narita 1964). …”

mentioning

confidence: 99%

Mimetic Parallels in Korean and Japanese

Garrigues

1995

View full text Add to dashboard Cite

Korean and Japanese are both known for their extensive utilization of mimetic adverbs. A comparative examination reveals the systematic nature of sound symbolism in the two languages and the striking parallels in their phonology and morphology. Similar mechanisms of symbolic sound alternation, suffixation and reduplication are utilized in both languages to extend the expressive range of mimetic words. The utilization of darkbright vowel pairs in Korean mimesis is well known, and a comparison between the two languages shows that Japanese also makes use of similar patterns of mimetic volume pairs. Not only does a comparative approach help illuminate the full systematic nature of sound symbolism in each language, but it may also help further our understanding of the wider relationship between the two languages.

show abstract

“…The implementation performed particularly poorly with the datasets bantudvd (Greenhill and Gray, 2015) (as illustrated by the 0.7053 B-Cubes for the 10% partition, versus 0.7835 of the Baseline submission), felekesemitic (Feleke, 2021) (0.6661 B-Cubes for the 10% partition, versus 0.6925 of the baseline). It performed better with datasets beidazihui (Běijīng Dàxué, 1962) (as illustrated by the 0.8356 B-Cubes for the 10% partition, versus 0.7279 of the baseline), bodtkhobwa (Bodt and List, 2022) (0.7993 B-Cubes for the 10% partition, versus 0.7566 of the baseline), bremerberta (Bremer, 2016) (0.7915 B-Cubes for the 10% partition, versus 0.7187 of the baseline), deepadungpalaung (Deepadung et al, 2015) (0.8143 B-Cubes for the 10% partition, versus 0.7597 of the baseline), wangbai (Wang and Wang, 2004) (0.8326 B-Cubes for the 10% partition, versus 0.8048 of the baseline), hattorijaponic (Hattori, 1973) (0.8127 B-Cubes for the 10% partition, versus 0.7889 of the baseline), listsamplesize (List, 2014) (0.5325 B-Cubes for the 10% partition, versus 0.4048 of the baseline). Full results are available along with the submission, with performance for other datasets comparable to the baseline.…”

Section: Implementation and Resultsmentioning

confidence: 99%

Approaching Reflex Predictions as a Classification Problem Using Extended Phonological Alignments

Tresoldi¹

2022

Preprint

View full text Add to dashboard Cite

This work describes an implementation of the "extended alignment" (or "multitiers") approach for cognate reflex prediction, submitted to "Prediction of Cognate Reflexes" shared task. Similarly to List et al. forthcoming(b), the technique involves an automatic extension of sequence alignments with multilayered vectors that encode informational tiers on both site-specific traits, such as sound classes and distinctive features, as well as contextual and suprasegmental ones, conveyed by cross-site referrals and replication. The method allows to generalize the problem of cognate reflex prediction as a classification problem, with models trained using a parallel corpus of cognate sets. A model using random forests is trained and evaluated on the shared task for reflex prediction, and the experimental results are presented and discussed along with some differences to other implementations.

show abstract

Japanese Dialects

Cited by 10 publications

References 0 publications

Automatic Inference of Sound Correspondence Patterns across Multiple Languages

Automatic Inference of Sound Correspondence Patterns across Multiple Languages

Mimetic Parallels in Korean and Japanese

Approaching Reflex Predictions as a Classification Problem Using Extended Phonological Alignments

Contact Info

Product

Resources

About