Cambridge Research Proposal Example: Forensic linguistics researcher to evidence policy (Score 93)

Forensic linguistics occupies an unusual position in English criminal courts: its outputs — authorship attribution analyses, speaker profiling reports, disputed-document assessments — are regularly tendered as expert evidence, yet no statutory or regulatory instrument sets minimum methodological standards for their admission. The Criminal Procedure Rules 2020 and the Law Commission's 2011 report on expert evidence both gesture towards reliability gating, but neither addresses the discipline-specific question of what counts as a sufficiently validated linguistic method for courtroom use. The result is an admissibility regime that is formally neutral but substantively uneven: courts evaluate forensic linguistic evidence through the same generic expert-evidence framework applied to forensic pathology or digital forensics, without the discipline-specific benchmarks that those fields have, over time, accumulated. This proposal asks: what regulatory mechanism — whether judicial, statutory, or professional-body-led — would most effectively and proportionately govern the admissibility of forensic linguistic evidence in English criminal proceedings, and what methodological criteria should that mechanism require practitioners to satisfy? Two subsidiary questions follow. First, to what extent do current admissibility decisions track the methodological quality of the linguistic analysis actually presented? Second, how have comparable common-law jurisdictions — specifically Australia and Canada — addressed the same regulatory gap, and what lessons, if any, transfer to the English context? The practical stakes are not abstract. Wrongful convictions in which disputed-authorship or voice-comparison evidence played a role have been documented in appellate case law; equally, methodologically sound forensic linguistic analysis has been excluded or discounted because courts lacked the evaluative vocabulary to assess it. A clearer regulatory framework would serve both defendants and the integrity of the fact-finding process. Two bodies of scholarship bear on this question but have not been brought into sustained dialogue. The first is the forensic linguistics literature itself. Coulthard, Johnson, and Woolls have mapped the range of analytical tasks the discipline performs in legal contexts, and Solan and Tiersma's work on language and the law provides a foundational account of how linguistic evidence functions in adversarial proceedings. More recently, researchers including Grant and Morrison have examined the validation problem directly, arguing that many authorship attribution methods lack the error-rate data that Daubert-style reliability standards would require. This literature is technically sophisticated but tends to treat the law as a fixed constraint rather than as a variable that could itself be redesigned. The second body of work is the law-and-regulation literature on expert evidence reform. Here, scholarship by Edmond, Roberts, and Mnookin has examined how courts in England, the United States, and Australia have struggled to operationalise reliability standards without the institutional capacity to conduct independent scientific review. The Law Commission's 2011 proposals for a reliability-based admissibility test were not enacted, and subsequent commentary has debated why the common-law gatekeeping model has proved resistant to legislative codification in England. This literature is analytically precise about the regulatory options but rarely engages with the specific methodological features of any individual forensic discipline. The gap this project addresses sits at the intersection: there is no sustained, empirically grounded analysis of how English courts have actually handled forensic linguistic evidence across a defined period, nor any comparative regulatory analysis that draws on that empirical base to evaluate which governance mechanism is best suited to the discipline's particular characteristics. Existing proposals for expert evidence reform are discipline-agnostic; existing forensic linguistics scholarship does not translate its methodological concerns into regulatory design terms. The project proceeds in three phases, each producing a discrete output that feeds the next. Phase one is a systematic case-law analysis. Using the BAILII database and, where necessary, Westlaw UK, I will identify all reported English and Welsh criminal cases from 2000 to 2024 in which forensic linguistic evidence was tendered, challenged, or discussed in the judgment. Preliminary scoping suggests a corpus of approximately 60 to 90 cases, a manageable number for close doctrinal analysis within the MPhil timeframe. Each case will be coded against a structured protocol covering: the type of linguistic analysis presented (authorship attribution, speaker comparison, discourse analysis, or other); the basis on which admissibility was contested or accepted; whether the judgment engaged with the methodology's validation status; and the outcome of the challenge. This phase produces both a descriptive account of current judicial practice and a dataset for the comparative phase. Phase two is a comparative regulatory analysis. Drawing on equivalent case-law searches in Australian and Canadian databases (AustLII and CanLII respectively), and on secondary literature covering those jurisdictions' expert evidence frameworks, I will examine how each system has responded to the methodological validation problem in forensic linguistics specifically. The comparison is deliberately narrow: I am not attempting a general comparative study of expert evidence law, but asking the focused question of whether either jurisdiction has developed discipline-specific admissibility criteria for linguistic evidence and, if so, through what institutional mechanism. Phase three synthesises the findings into a regulatory design analysis. Drawing on the empirical case-law data and the comparative material, I will evaluate three candidate regulatory mechanisms — judicial reliability gating, a statutory admissibility standard with discipline-specific criteria, and accreditation through a professional body such as the International Association of Forensic Linguists — against criteria of proportionality, institutional feasibility, and methodological adequacy. The output is a policy-oriented analysis, not a draft statute, and the contribution is explicitly bounded: I am not claiming to resolve the broader expert evidence reform debate, but to provide the discipline-specific empirical and comparative grounding that that debate currently lacks. All primary data sources — BAILII, Westlaw UK, AustLII, CanLII — are publicly accessible, and the Faculty of Law's library subscriptions cover the required databases. No human participants are involved, so the project does not require ethics committee approval beyond standard research registration. The main feasibility risk is corpus size: if the BAILII search returns fewer than 40 usable cases, the descriptive phase will be thinner than planned, and I will supplement it with tribunal and appeal decisions not captured in standard criminal case-law searches. Conversely, if the corpus exceeds 100 cases, I will apply a principled sampling protocol — prioritising appellate decisions and cases in which admissibility was directly contested — to keep the analysis tractable. Provisional timeline: Michaelmas term — systematic case-law search, coding protocol development, and initial literature mapping. Lent term — case-law coding completed, comparative analysis of Australian and Canadian material, and a draft of phases one and two. Easter term and summer — regulatory design analysis, full draft, and revision for submission. This schedule is consistent with the MPhil's one-year structure and builds in a contingency buffer of approximately three weeks before the submission deadline. The Faculty of Law at Cambridge has active research interests in evidence law, regulatory theory, and the intersection of science and legal process. The Centre for Corporate and Commercial Law and the work associated with the Cambridge Regulatory Toolkit both address questions of regulatory design that are directly relevant to phase three of this project. The Faculty's evidence law scholars have published on expert evidence reform and on the epistemological challenges courts face when evaluating scientific testimony — the precise doctrinal terrain this project occupies. The project also connects to the Faculty's broader interest in empirical legal research methods. The case-law coding methodology in phase one draws on techniques developed in empirical legal studies, and the MPhil in Law and Regulation's emphasis on research design and regulatory analysis provides the methodological training the project requires. The linguistics dimension of the work is grounded in my undergraduate training in authorship attribution and corpus methods; the regulatory and doctrinal analysis is the new intellectual territory the MPhil is designed to develop. The combination is, I think, the right fit for a one-year research degree: the methods are within my existing competence, the legal-regulatory framing is the extension, and the question is bounded enough to be completed within the programme's timeframe.