Abstract
Forty years ago, Vaughn Ball demonstrated that the then received notion of conditional relevance served no useful purpose, as it would only come into effect if the probability of an element were 0.0. But, if the probability of an element were 0.0, a directed verdict would be in order and so once again conditional relevancy was doing no work. I extended that analysis to include the relationship between proffers of evidence and facts of consequence to demonstrate that the work that conditional relevancy was supposedly doing was isomorphic to that done by relevancy in all significant ways, and yet the Federal Rules of Evidence provide different standards for the two situations to determine admissibility. I thus proposed an amendment to FRE 104(b) to provide for the same standard to be applied to ‘relevancy’ and ‘conditional relevancy’. In a recent article, Matthew Kotzen appears to subject this work to intense scrutiny and criticism, yet at the same time reaches, so far as I can tell, identical conclusions. This raises the question, which I examine, whether this is an example of cross-disciplinary difficulties in communication.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Almost all of the work published in English, including that from scholars from across the world, focuses on the American legal systems, as I will here as well. On the uselessness of weird hypotheticals, see Ronald J. Allen, Naturalized Epistemology Revisited, 2 Int. J. on Evidential Legal Reasoning: Quaestio Facti 1–32 (2021).
Ronald J. Allen & Chris Smiciklas, The Law’s Aversion to Naked Statistics and Other Mistakes, Legal Theory, 28(3): 179–209 (2022).
See for example the responses to Allen, supra n. 1, in Vol 3 of the Int. J. on Evidential Legal Reasoning: Quaestio Facti (2022).
Matthew Kotzen, Conditional Relevance and Conditional Admissibility, Law and Philosophy 42: 237–283 (2023), hereinafter referred to as Kotzen.
The probabilistic model that Kotzen deploys to analyze relevance is an inaccurate model of American evidence law and trials. Probabilistic modeling of juridical proof has given way to explanationism for various reasons. For example, most admissible evidence is not ‘relevant’ under the traditional view as evidence at trial is typically massively overlapping. See Ronald J. Allen & Michael, Pardo, Relative Plausibility and Its Critics, 23 Int. J. Evidence and Proof 5–58 (2019), and the twenty responses to that article. The questions at trial are, first, the relationship between a proffer and a party’s explanation, not whether it increases or decreases the probability of various propositions in the abstract, and second, is the fact finder able rationally to appraise the evidence. A useful contribution would be to work through the prior analyses from the perspective of this new lens. See e.g. Michael Pardo, What Makes Evidence Sufficient, forthcoming, 65 Arizona L. Rev., available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4437441&dgcid=ejournal_htmlemail_criminal%3Aprocedure%3Aejournal_abstractlink. I put these complexities aside for the purposes of this reply.
Kotzen refers to this as ‘unconditionally relevant’, or ‘logically relevant’. Nothing is ‘logically’ or ‘unconditionally relevant’ at trial; evidence is inert unless appraised by a cognizer on the basis of that person’s knowledge and beliefs. See, e.g., Ronald J. Allen, Factual Ambiguity and a Theory of Evidence, 88 Nw. U. L. Rev. 604, 627 (1993–1994) (implications of trial observations are ‘determined by the sum total of that person’s experiences at the moment of decision, experiences which will by that time include the advocates’ efforts to enlighten the fact finder about the implications of the material produced at trial and all the other observations generated by the trial’). Although he recognizes the misleading nature of these labels in that relevancy is conditional upon what a reasonable person might know at the time, Kotzen, pp. 237–240, his use of the labels nonetheless leads him astray. He asserts, for example, that ‘Whether an item of evidence is logically relevant does not change as more evidence is introduced into the record, since the relevance of logically relevant evidence does not depend on other evidence in the record’. Kotzen, p. 242. This statement is true to a logician, but false in the context of trials. Valid deductions from assumptions do not change in formal logic, but Kotzen neglects that at trial the ‘assumptions’ are always open to contest. A common phenomenon at trial involves ‘other evidence’ that shows some supposed ‘common-sense’ belief likely to be false which raises doubts about the admissibility of the evidence premised on the now doubtful common-sense belief. He notes an aspect of this in n. 14, but mistakenly thinks that ‘this complication will not impact my aims in this paper, in large part because the undercutting evidence at issue would typically be introduced after the logically relevant evidence and hence would not interfere with its admissibility’. He seems unaware of subsequent motions to exclude and disregard the evidence. There are other examples of unfamiliarity with the realities of evidence law and trial practice. See n. 30, infra.
The only interesting difference is the standards of proof. Obviously, if a judge does not think a juror (or the judge) can appraise the evidence as it stands, further evidence will have to be offered. That can be characterized as a ‘procedural’ difference, but it is trivial. The significant question is why the standards for admissibility differ depending on whether a factfinder’s knowledge and experience prior to trial extends to the intermediate premises needed to draw the connection between an evidentiary proffer and a fact of consequence or whether the proponent of evidence needs to adduce evidence of those intermediate propositions. Kotzen and I agree that the standards should not differ. Indeed, that seems to be the primary point of both his and my work, which makes the extensive criticism in Kotzen’s paper difficult to comprehend.
It is contained in the first chapter of Allen, Swift, Schwartz, Pardo, and Stein, An Analytical Approach to Evidence: Text, Problems, and Cases, 6th ed. (2016).
Ronald J. Allen, The Myth of Conditional Relevancy, 25 Loyola L. Rev. 871–884, 879 (1992).
Id. at p. 883.
Id. at , p. 882.
His paper goes off in other directions as well. See n. 27, infra.
Kotzen, p. 258.
Although I am happy to credit Kotzen with this point in the sense that he is right that it is missing from my original paper, in fact it is such an obvious point that it is contained in the coursebooks on evidence. See, e.g., Allen et al., supra n. 8 at pp. 134–144. We also point out that an analogous issue pertains to determinations under FRE 403, which is a part of Kotzen’s analysis that is not pertinent to the present discussion. Id. at pp. 152–155.
See, e.g. California Code, Code of Civil Procedure—CCP § 631.8.
Kotzen, p. 256.
Id. at 882.
Allen, supra n. 9 at 883–884.
Allen, supra n. 6, at p. 608, introduced these logical curiosities into the evidence literature.
Kotzen, p. 247.
Id. My argument also did not depend on the distinction between evidence increasing or decreasing the probability of a fact of consequence.
See FRE 401.
It beggars belief to suggest that Ball was unaware that evidence negating an element or a combination of elements is relevant, yet that is what Kotzen’s argument apparently depends upon. Kotzen, in a bit of presentism, is taking Ball to task for employing the then conventional modes of discourse surrounding this issue. He presented the problem in its conventional mode to correct a misunderstanding about it.
Kotzen, p. 250.
In a mystifying passage elaborating on why the murder hypothetical shows that relevancy and conditional relevancy do not collapse into each other, because evidence of the dominant hand of the killer is relevant conditional on evidence of the dominant hand of defendant, Kotzen says at p. 250 ‘This example suffices to show, in a second way, that there are non-trivial cases in which evidence is conditionally relevant and yet not relevant. And again, the general lesson is conditional relevance does not collapse into relevance; they are most emphatically not the same concept’. This is simply another example of where an intermediate premise is needed, and the issue is the standard to be applied in determining whether evidence of it is adequate, whether in the minds of the fact finder or in the record evidence. Kotzen seems to be asserting that I denied such a situation exists, when in fact it was the very focus of the article.
FRE 103.
For example, in Ondine Shipping Corp. v. Cataldo, 24 F.3d 353 (1st Cir. 1994), over the district court’s suggestion that the plaintiff was making a mistake, the plaintiff nonetheless pursued an unconventional method of proving damages, which ultimately was to the plaintiff’s detriment. Master of the case doctrine is essentially the law everywhere within the U.S. and the cases are ubiquitous. I and some colleagues are preparing a manuscript that discusses such matters, which is entitled Minimal Rationality and the Law of Evidence, which we would be delighted to share.
Putting aside plain error, which is not what we are arguing about. See FRE 103.
See n. 1, supra. This problem arises because, as in some of Kotzen’s hypotheticals, there is the possibility of multiple different inferences. That may raise a problem of undue prejudice but not of relevance. Even if the inferences cancel each other out in some fashion, the evidence will be admissible because it helps one party in a certain fashion and the other in some other fashion. The parties get to present their theory of the case as they like. See n. 5, supra.
See n. 6, supra. There are other examples. No one responsibly trained in American evidence law and at all observant of its application can fail to notice that trials involve ‘conditional’ evidence from beginning to end, including the recurring need to connect things up: ‘What any offer of evidence means depends on all the evidence in the case. What might appear initially to be inculpatory may actually turn out to be exculpatory’. Allen & Pardo, supra n. 5 at p. 12, n. 43. There are other peculiar components to the analysis that demonstrate a lack of understanding of how American evidence law operates, no matter how philosophically creative the analysis might be. Perhaps the paradigmatic example is his argument:
consider a situation in which A and B are each unconditionally relevant and are totally unrelated to each other, and further suppose that while A easily satisfies 403 analysis by itself, B just barely fails 403 analysis by itself. In this situation, it is clearly appropriate for the court to admit A and to refuse to admit B. However, if the proponent were to persuade the court to treat A and B as a single evidential package, that package would pass a 403 analysis; A’s easy passage would more than make up for B’s narrow failure, leading to a package in which net probative value is not substantially outweighed by net danger of unfair prejudice. Thus, if the court were to treat A and B as a single package, the result would be for B to be effectively ‘smuggled’ into evidence on A’s evidentiary coattails.
How could a proponent ‘persuade the court to treat A and B as a single evidential package’, when they obviously are not? What does that even mean? That the judge could be convinced to consider a couple of unrelated witnesses or lines of questioning and average the various results? The suggestion that trial judges might behave in this fashion is preposterous.
He also argues for combining the relevance determination with a probative value/prejudicial effect analysis, which is beyond the scope of my interest.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Allen, R.J. Kotzen, Conditional Relevancy, and the Difficulties of Cross-Disciplinary Dialogue. Law and Philos 43, 215–225 (2024). https://doi.org/10.1007/s10982-023-09489-7
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10982-023-09489-7