A Philosophy of Evidence Law

TL;DR

This chapter argues that the standard of proof is better understood from an internal perspective as a “standard of caution” rather than as a fixed decisional probability threshold. It distinguishes the degree of caution, which should vary with the seriousness of the allegation and its consequences, from the distribution of caution, which differs in kind between civil impartiality and criminal protection of the accused.

Key findings

• The traditional external analysis of standards of proof has three problematic features: it focuses on the end-state of belief, treats the standard as a fixed decisional threshold, and evaluates it as an instrument of social policy. The chapter rejects each feature and proposes an internal, process-oriented account.
• The proposed variant standard holds that the fact-finder must find that *p* only if justified in believing sufficiently strongly that *p* within the terms of BAF*; the required confidence is context-dependent and increases with the gravity of the allegation and the consequences of accepting it. The chapter cites judicial support for a flexible civil standard, including Lord Denning’s statement that “In some cases 51 per cent would be enough, but not in others,” and notes that courts reject attempts to quantify “beyond reasonable doubt” by percentages.
• The standard has two aspects: degree of caution and distribution of caution. In degree, there is no difference in kind between civil and criminal cases—there should be one variant standard. In distribution, the civil standard requires impartiality and equal treatment of both parties, while the criminal standard requires a protective, asymmetrical attitude toward the prosecution, under which the fact-finder must acquit if a reasonable doubt is cast on guilt even without positively believing the defence hypothesis.
• The asymmetry is grounded in the moral judgment that it is worse to convict the innocent than to acquit the guilty, and in justice as empathic care and respect for the person affected by the finding. The chapter says the moral judgment’s precise justification is difficult to articulate.

Methods

• Conceptual and normative legal-philosophical analysis; no empirical sample, data, or quantitative findings are reported.
• The chapter critiques the external/dominant understanding of standards of proof, then develops an internal perspective drawing on epistemology—especially epistemic contextualism and interest-relative accounts of knowledge/justified belief—and on an ethics of care.
• It engages legal authorities and hypotheticals (e.g., *Briginshaw v Briginshaw*, *Bater v Bater*, *Re Winship*, European Court cases) alongside philosophical literature (e.g., Cohen, Wedgwood, Stanley, Hawthorne, Fantl and McGrath, Weatherson).

Limitations

• The paper states that the moral judgment that it is worse to convict the innocent than to acquit the guilty is difficult to justify precisely; it offers “only a few words” on that justification and notes that empirical support is lacking for the assumed connection between moral gravity and frequency of conduct.
• The paper acknowledges possible objections to a variant standard—that it licenses carelessness in less serious cases, ties proof to extra-epistemic policy, creates uncertainty, or violates equality—and attempts to rebut them.
• Evident from the text: the argument is normative and conceptual rather than empirical, relies on selected common-law authorities and philosophical literature, and does not test the proposed standard against systematic data.