多模态语步分析:以微信科普文为例

Move Analysis of Popular Science Articles on WeChat:A Multimodal Perspective

本文采用Kress和van Leeuwen视觉语法中的组篇功能,以及Martinec和Salway的图文逻辑语义关系,提出了多模态语步分析方法,即自上而下的语步识别和自下而上的语步实现。本文具体研究了语篇语步结构如何由信息值、凸显性以及框架等多模态资源识别;以及语言和其他模态资源如何通过详述、增强以及延伸关系共同实现语步。本文将信息值重新定义为宏主位/述位;宏主位构成语义信息的起始点,与上一语篇片段区分。宏主位/述位,凸显性以及框架三者共同区分不同的信息结构,揭示交流目的转变,帮助识别具有不同交际目的的语步。融媒体时代下,Martinec和Salway的图文逻辑语义关系可扩展为语言与非语言符号间的逻辑语义关系,这些关系具有复杂性和重叠性特征。最后,以微信公众号“丁香医生”发布的70篇新冠肺炎科普文为例,验证此多模态语步分析方法的实用性。

Drawing upon Kress and van Leeuwen’s visual grammar,in particular,the compositional metafunction and Martinec and Salway’s logico-semantic relations for image and text,proposed a model for multimodal move analysis,that is,top-down approach to move identification and bottom-up approach to move realization.Particularly,the present study investigated how Information Value,Salience,and Framing contribute to move identification and how elaboration,enhancement and extension relations contribute to move realization.The Information Value is re-positioned as Macro-Theme and Rheme in the current study,which initiates the departure of new information.Macro-Theme/Rheme,Salience and Framing differentiate different move types with different communicative purposes.Martinec and Salway’s image-text semantic relations are extended to verbal and non-verbal relations characterized with complexity in multimedia age.Finally,we used a total of 70 science popularization on WeChat to illustrate the framework’s value.