Five Languages of Eurasia: Field Work, Analysis and Digital Archiving

该项目由NSF记录濒危语言计划和北极社会科学计划资助，由高露洁大学的Alexander Nakhimovsky博士和莫斯科大学Alexander Kibrik博士领导的一个俄罗斯语言学家小组领导，该项目将实施一个为期三年的项目，重点关注两个领域：(1)关于五种语言的实地工作：高加索地区的阿奇语和吉纳卢格语，泰米尔半岛的Nganasan和Enet语，堪察加半岛的Alutor语；以及(2)开发创建语言档案的工具和基础设施。第一个领域的工作将由俄罗斯学者完成，他们是所涉及语言的顶级专家，但缺乏技术和专业知识，无法按照“最佳做法”创建语言档案。第二个领域的工作将由PI完成，他是一名语言学博士，自1985年以来一直在教授计算机科学，教授并广泛出版了这些“最佳实践”所基于的XML和语义网技术。该项目的参与者将：--开展广泛的实地工作，记录所有语文的数字音频和视频材料；--按照最佳做法进行修订、数字化和记录，并提供更广泛的查阅以前收集的材料的途径，可追溯到1970年代；--设计两种语文的书写系统和其中一种语文的教育材料；--为三种语文创建一个语音数据库，其中包含从音节到单词、从句子到叙事的不同级别的声音样本，并配有显示发音能力的视频；-根据数字语文档案的最佳做法编排原始材料的格式，并以一致的、可存档的、可互操作的和基于网络的格式将其作为动态的OLAC储存库提供；-进一步发展基础设施和软件工具，以便在创建语文档案方面的“最佳做法”，包括PI为数字视频添加注释的独特软件工具。这项提案将利用顶尖学者的专业知识和正在进行的研究，他们将在此提案的边界之外，在整理和描述数据方面贡献大量劳动密集型和专业知识密集型工作。他们将致力于记录濒危或濒临灭绝的语言，在极其偏远或处于政治动荡中的地区，在他们已经拥有了解和信任他们的线人网络的社区。此外，该提案还将制作软件工具，用于开展实地工作和在网上存档其结果。其中一些将是从领域专家那里收集元数据的非常通用的工具。该提案完成后，将使几个社区受益。首先，参与该项目的一些以英语为母语的社区将获得书写系统和儿童教育材料。其次，俄罗斯语言学家社区将得到强大的推动力，促使他们采取更好的档案做法。他们中的许多人都是训练有素、积极进取的人，他们多年来一直在研究俄罗斯联邦的许多濒危语言。现有的这些关于他们工作的文件大多是纸质的或专有格式的，用俄语写成，录音带缺失或质量低下。如果他们的文献标准得到提高，世界各地的语言学家和人类学家将获得妥善存档的大量语言知识。第三，世界各地的语言学家社区，不仅仅是那些致力于濒危语言的人，将获得强大的新工具，其中一些工具捆绑到Sil的流行工具箱中进行实地考察。最后，该项目将使正在构建语义网的计算机科学家和技术人员受益。语义网是一个由记录良好、索引良好的信息组成的网络，可以支持复杂的搜索、导航和推理。

This project, which is funded through the NSF Documenting Endangered Languages program and the Arctic Social Sciences program, led by Dr. Alexander Nakhimovsky of Colgate Univiersity and a group of Russian linguists led by Dr. Alexander Kibrik of Moscow University will conduct a three-year project that will focuse on two areas: (1) field work on five languages: Archi and Khinalugh in the Caucasus, Nganasan and Enets on the Taimyr peninsula, Alutor on Kamchatka; and (2) development of tools and infrastructure for creating language archives. Work in the first area will be done by Russian scholars who are top experts on the languages involved but lack technology and expertise to create language archives in accordance with "best practices." Work in the second area will be done by the PI, a PhD in linguistics who has been teaching computer science since 1985 and has taught and extensively published on XML and Semantic Web technologies on which these "best practices" are based. Participants of the project will: - conduct extensive fieldwork to record digital audio and video materials in all languages; - revise, digitize, and document in accordance with best practices, as well as provide wider access to previously collected materials, going back to the 1970s; - devise a writing system for two languages, and educational materials for one of them; - for three languages, create a phonetic database containing sound samples at different levels, from syllable to word to sentence to narrative, with videos showing articulation; - format primary materials in accordance with best practices for digital language archives and make them available, as dynamic OLAC repositories, in consistent, archivable, interoperable, and Web-based formats; - further develop the infrastructure and software tools for "best practices" in creating language archives, including PI's unique software tool for annotating digital video. This proposal will leverage the expertise and ongoing research of top-notch scholars who will contribute, beyond the boundaries of this proposal, a great deal of labor- and expertise-intensive work on glossing and describing the data. They will work to document languages that are either endangered or on the brink of extinction, in areas that are either extremely remote or in political turmoil, in communities where they already have networks of informants who know and trust them. In addition, the proposal will produce software tools both for conducting fieldwork and for archiving its results on the Web. Some of them will be quite general tools for collecting metadata from domain experts. When completed, the proposal will benefit several communities. First, some of the communities of native speakers involved in the project will receive a writing system and educational materials for children. Second, the community of Russian linguists, many of them highly trained and motivated people who have been working on the many endangered languages of the Russian federation for years and decades, will receive a powerful impetus to adopt better archival practices. Such documentation of their work that exists now is mostly on paper or in proprietary formats, in Russian, with audiotapes missing or of poor quality. If their documentation standards improve, linguists and anthropologists worldwide will get access to vast linguistic knowledge properly archived. Third, the community of linguists everywhere, not only those working on endangered languages, will receive powerful new tools, some of them bundled into SIL's popular toolbox for fieldwork. Finally, the project will benefit computer scientists and technologists who are building the Semantic Web: a web of well-documented and well-indexed information that can support complex search, navigation and inference.