Dillon on Scientific Expertise

James R. Dillon (Columbia Law School, Associate in Law) recently posted an article titled Expertise on Trial to SSRN. Here is the abstract:

The problem of “epistemic competence” — the inability of courts to effectively interpret and apply scientific expert testimony to the resolution of legal disputes — has been a vexing one nearly as long as expert witnesses have been routine fixtures in litigation. Some have despaired that the problem seems intractable; while scientific expert testimony has come to play an increasingly prominent role in a broad range of litigation, we are no nearer a solution to the problem of epistemic competence than when the conversation began in the nineteenth century.

This Article argues that the intractability of the problem is at least in part the result of the terms under which the discussion has been framed. The epistemological paradigm of the existing literature makes an impossible demand: that individual legal decision makers possess substantive expertise in all scientific fields in which expert witnesses testify. Because judges and jurors are not omniscient, this demand can never be satisfied, and reform proposals have therefore been limited to mitigating the problem of epistemic competence with no hope of solving it.

This Article proposes a new solution to the problem of epistemic competence. Drawing on social epistemology and the sociology of scientific knowledge, it advocates a collectivist epistemological paradigm wherein the institution of the court, rather than the individual judge and jurors, is the epistemic agent of interest. The Article describes a system of distributed cognition that would vest scientific expertise in courts as institutional epistemic agents, thus solving the problem of epistemic competence. That system is not without cost, however, as the necessary reforms would require substantial departures from existing judicial structures. Thus, the Article ends with a dilemma: either courts must substantially reform their institutional structures to gain competence in evaluating scientific testimony, or we must reframe courts’ institutional mission to match their epistemic capacities.

In a nutshell, Dillon proposes to solve (or at least minimize!) certain epistemic problems facing courts by creating an "Office of Scientific Adjuncts," which would provide disinterested scientific experts to help judges evaluate scientific evidence more accurately and efficiently. While I suspect his proposal faces long odds indeed, it provides an interesting framework in which to think about the epistemic problems facing courts and how they could be addressed more effectively. Dillon presented a draft version of the paper earlier this year in the faculty lecture series at the University of Kentucky College of Law, and I'm happy to see the final version available.

Comments

[Patrick S. O'Donnell]

I've yet to carefully read the paper (hope to anon), but finding "disinterested scientific experts" (say, in the minimal sense of immune from commercial or predominantly personal 'professional' interests) is no easy (although not impossible) task these days, given the nature of contemporary scientific praxis. Skimming through the notes/references, I did not see any mention of Susan Haack's work (perhaps I missed it), which I believe is eminently relevant to this topic. In particular, see her Evidence Matters: Science, Proof, and Truth in the Law (CUP, 2014). As Haack points out,

1. In the scientific fields or specialties germane to toxic court cases, the research that bears on issues of causation "is likely to have been conducted by a drug company or a chemical manufacturer ... quite often, with an eye to protecting itself against litigation--exactly the kind of scientific work whose commercial interests severely strain the informal scientific mechanism that encourage honesty and discourage the withholding of evidence."
2. "The cases that come to trial will normally be those where the evidence is ... far incomplete and ambiguous," a situation in tension with the legal aspiration to "resolve disputes promptly," and thus scientific expertise as such will not dissolve or transcend this tension.
3. "Because of its case-specificity, the law often demands answers of a kind that science is not well-equipped to supply." She cites the example of Joiner v. General Electric Co. 864 F. Supp. 1310 (N.D. Ga. 1994) ['I-IV']
4. "The legal system constitutes virtually the entire market for certain fields of forensic science (or quasi-science), and for certain psychiatric specialties."
5. "Legal rules can make it impossible to bring potentially useful scientific information to light." Perhaps here Dillon's proposal may be capable of altering the epistemic dynamics.
6. "The legal penchant for rules, 'indicia,' etc. sometimes transmits scientific subtleties into formulaic legal shibboleths." This appears to be a fairly obdurate yet not necessarily insoluble problem.
7. "Because of its concern for precedent, and the value it places on finality, the legal system has a tendency to inertia, and sometimes lags behind science." Here again Dillon's suggestions may make some difference.

[James Dillon]

Patrick,

Thank you for your comment. The issues you raise are important ones; I do briefly attempt to address them in Part IV(B)(3)(iii) of the paper, a discussion that I doubt you'll find fully satisfactory (nor do I) as I have space there to only briefly acknowledge these broad issues about which much more can be said. I don't cite Haack in that section (I do have one passing reference to her work at footnote 44, and will take a look at the article you cite as it does sound quite relevant), but Mnookin and Jasanoff have made similar observations. I would say two things: 1) the evidentiary regime under FRE 702/Daubert does assume a sort of naïve positivism that those who study the philosophy or sociology of scientific knowledge will recognize as incomplete (I think of this as a reification of propositional knowledge that scientists themselves would understand to be necessarily provisional; the "formulaic legal shibboleths" to which Haack refers), and for the most part, for purposes of this paper, I'm working within that paradigm rather than attempting to subvert it simply because *even if* we accept the law's naïve positivism, existing institutional structures still are not very good at incorporating that body of propositional knowledge into the legal decision making process; and 2) I do think that, insofar as my solution calls for bringing substantive scientific experts directly into the legal decision-making process, it will have the effect of mitigating some of the law's naiveté about the nature of scientific knowledge; that isn't a complete solution because, as you point out, individual experts have their own personal and professional biases, but I would call it an improvement on current practice.

I hope in the foreseeable future to turn more directly to the problem of the mismatch between scientific conceptions of knowledge and the demands, in terms of certainty, finality, and the speed at which knowledge can be generated, that the law places on scientific expertise; that's an entry on my current research agenda though I expect it to be 2-3 papers down the line, as I'm still trying to think through problem myself. I suspect that it will prove to be one that can only ever be mitigated rather than fully resolved, and that the path to mitigation lies in changing the rules of civil procedure to accommodate the provisionality of scientific consensus-building while nevertheless respecting the law's need for finality and reasonably prompt resolution of legal disputes.

James

[Patrick S. O'Donnell]

Thanks for your thoughtful reply James. (The Haack material was from a chapter in the book, perhaps its first incarnation was as an article, I don't know.) I do look forward to reading your paper with the care it deserves as well as anything else you write on law and science. And I'm grateful to Brian for letting us know of your work.