Domestic Abuse in the Courtroom: A Corpus-Aided Discourse Analysis of the Construction of Victimhood at Trial

据估计，在截至2022年3月的一年中，近10%(9.9%)的英格兰和威尔士成年人经历过家庭暴力(以下简称DA)。尽管这一社会问题非常突出，但关于其在试验环境中的语言表征的研究却很有限。因此，本项目结合了法医语言学、语料库语言学和语篇分析的要素，以德普诉希尔德案(2019)诽谤审判为案例，探索发展议程在法庭上的语言表征。尽管这场审判的法律问题涉及安珀·希尔德在一篇专栏文章中陈述的真实性，但此案取决于是否发生了虐待行为，以及随着审判的展开，谁是受害者：德普还是希尔德。检察官故事对案件的中心地位使其成为解决该项目研究目标的适当数据来源；这些故事特别侧重于指控检察官的证人如何在法庭上引导受害者的语言，以及这一过程如何由他们的审判律师促进和禁止。在法庭强奸的背景下，有大量文献证明基于性别的暴力的受害者在证明其陈述时面临的挑战。已有证据表明，交叉审讯者使用语言手段，如重新表述和控制问题类型，来支配强奸审判中的互动空间，并歪曲申诉人的证词(Drew 1992；Ehrlich 2001)。这些装置将犯罪叙事转变为将导致强奸的行为归咎于受害者，并表明他们实际上希望性交发生(Taslitz 1999；Matoesian 2001；Ehrlich 2010)。关于DA报纸代表性的研究(例如Lindsay-Brisbin等人)2014)表明这种形式的受害者刻板印象在媒体话语中也很普遍；诸如指责受害者，攻击受害者的性格，将他们描述为乱交，以及通过将犯罪者描述为具有无法控制的愤怒来为犯罪者开脱，这些主题都被证明在英国和美国的报纸上长期存在(Taylor 2009；Braber 2014；Lloyd and Ramon 2017)。由于媒体在塑造社会问题的公众舆论方面拥有巨大的影响力(Tuchman 1978；Higgins 2010)，而且检察官审判中的陪审员是公众的代表，本研究将探索这些主题在多大程度上渗透到法庭话语中。该项目将采取语料库驱动的探索性方法，依靠从数据中出现的模式来指导分析的方向(Togniti-Bonelli 2001)。为了促进这一点，将创建一个小型、专门的语料库，其中包括对德普和希尔德的直接和交叉询问；这些总共价值约50个小时的数据来自CourtTV(2023年)对审判的转播。搭配、频率列表和和谐的语料库工具都被有效地用于法律语篇的分析(Heffer 2005；Potts and Weare 2018；Potts and Formato 2021)；因此，本项目将利用这些工具对语料库进行词汇分析，研究发展议程及其围绕受害者的主题是如何在法庭上进行话语构建的。这项研究还将利用对话分析和语料库方法，这是O‘Keefe和Walsh(2012)首创的一种协同方法，以定性的、近距离的能力检查这些法庭记录的摘录；这将使量化的语料库调查结果在其交互背景下得到进一步探索。这项研究将提供对投诉人在以话语形式构建虐待故事时可能面临的体制(语言)限制的见解。了解这一点的意义重大；通过确定这些制约因素，这项研究不仅将提高人们的认识，还将指明法律改革的方向，使发展援助受害者有最好的机会在法律系统中伸张正义。

Almost 10% (9.9) of adults living in England and Wales were estimated to have experienced domestic abuse (hereafter DA) in the year ending March 2022 (ONS 2022). Despite the prominence of this social issue, there is limited research on its linguistic representation in a trial setting. This project therefore integrates elements of forensic linguistics, corpus linguistics and discourse analysis to explore the linguistic representation of DA in the courtroom, using the Depp v. Heard (2019) defamation trial as a case study. Though the legal issues of this trial concerned the veracity of Amber Heard's statements in an op-ed, the case hinged on whether or not abuse occurred and, as the trial unfolded, who was the victim: Depp or Heard. The centrality of DA stories to the case make it an appropriate data source for addressing this project's research aims; these have a particular focus on how witnesses alleging DA linguistically navigate victimhood in court, and how this process is both facilitated and prohibited by their trial lawyers. Within the context of rape in the courtroom, there is a wealth of literature evidencing the challenges victims of gender-based violence face in proving their accounts. Cross-examiners have been shown to use linguistic devices, such as reformulations and controlling question-types, to dominate the interactional space in rape trials and distort complainants' testimonies (Drew 1992; Ehrlich 2001). These devices transform the crime narrative into one that blames the victim for their actions leading up to rape, and suggests they actually wanted the intercourse to happen (Taslitz 1999; Matoesian 2001; Ehrlich 2010). Research on DA newspaper representation (e.g. Lindsay-Brisbin et al. 2014) suggests this form of victim stereotyping is also prevalent in media discourse; themes such as victim blaming, attacking the victim's character in depicting them as promiscuous, and excusing the perpetrator by portraying them as having uncontrollable rage, have all been shown to be perpetuated in newspapers in the UK and US (Taylor 2009; Braber 2014; Lloyd and Ramon 2017). Since the media hold great power in shaping public opinion on social issues (Tuchman 1978; Higgins 2010), and jurors in DA trials are representative of the public, this research will explore the extent to which these themes permeate into courtroom discourse. This project will take a corpus-driven, exploratory approach, relying on the patterns that emerge from the data to guide the direction of the analysis (Tognini-Bonelli 2001). To facilitate this, a small, specialised corpus will be created of the direct and cross-examinations of Depp and Heard; these total around 50 hours' worth of data from CourtTV's (2023) broadcasting of the trial. Corpus tools of collocation, frequency lists and concordances have all been effectively used in the analysis of legal discourse (Heffer 2005; Potts and Weare 2018; Potts and Formato 2021); this project will therefore utilise these tools to undertake lexical analysis of the corpus, examining how DA and its themes around victimhood are discursively constructed in the courtroom. The study will also utilise Conversation Analysis alongside corpus methods, a synergy pioneered by O'Keefe and Walsh (2012), to examine extracts of those courtroom transcripts in a qualitative, close-up capacity; this will enable the quantitative corpus findings to be explored further in their interactional context. This study will provide insights into what institutional (linguistic) constraints complainants may face in discursively constructing stories of abuse. The implications of gaining this understanding are significant; through identifying these constraints, this research will not only raise awareness, but could also indicate directions for legal reform that would give DA victims the best chance at attaining justice in the legal system.