Working papers

Expert witnesses in civil proceedings are legally required to be independent. The current recusal mechanism relies on self-declaration and adversarial challenge — a slow, qualitative process with no quantitative standard for assessing the network proximity between an expert and the parties they serve. This paper applies Erdős-number methodology on academic co-authorship and supervisory graphs to demonstrate that the existing procedural standard is structurally inadequate, and proposes a formal network proximity threshold — grounded in scale-free network statistics — as a required component of expert appointment procedure across European civil proceedings.

Legal proceedings treat qualification as a proxy for expertise. The two are not the same thing: a credential certifies past education; expertise requires demonstrated performance within a specific technical domain at sufficient depth and recency. The paper identifies measurable signals that reveal the gap between formal qualification and actual domain expertise — operationalised as the Specialisation Distance Differential (SDD) — and proposes how those signals can be assessed before an expert is appointed, not after they have already testified.

"Bad faith" in civil procedure is typically treated as a subjective, motive-based finding that requires prolonged litigation to establish. This paper reframes bad faith as a structural condition that can be diagnosed from the mechanism design of the dispute itself. It introduces the Bad Faith Manufacturing Score (BFM), a scored composite that identifies the structural preconditions for bad-faith conduct in advance of an adjudicative finding — enabling early procedural triage rather than post-hoc characterisation.

This paper names three structural features that recur in concealment-of-condition disputes: the Loaded Gun (prospective liability created by an undisclosed defect), the Fiction of Uniformity (liability provisions that assign responsibility by activity without examining variation between actors), and the Gridlock Paradox (the double-bind that makes both action and silence costly). The paper develops the doctrinal category of the Affected Party — the third-party whose position is materially altered by concealment they had no role in — and operationalises the analysis in Phase IV of the Bad Faith Check.

Adapts Harry Frankfurt's philosophical distinction between lying and bullshitting — where the liar knows the truth and misstates it, and the bullshitter is indifferent to the truth — to civil procedure, arguing that a substantial share of contested factual assertions in litigation are bullshit in Frankfurt's sense rather than lies. The paper introduces the Self-Reporting Fallacy: the moment a positive factual assertion is contradicted by its own attached evidence, converting the party's own submission into the strongest available proof of its inaccuracy. This is the "PI liability switch" moment the Dispute Engine detects.