Introduction
Early in the first year of law school, students are likely to realize that facts are crucially important. But the law school curriculum is designed so as to make the process of legal factfinding almost invisible. The traditional first year courses focus on appellate cases and legal norms. The facts are givens. The standards of appellate review largely insulate factfinding by trial courts from examination by appellate courts. And casebooks focus on legal rules and largely exclude cases that focus on the factfinding process. The primary law school course that focuses on factfinding is Evidence, but in many versions that course, the emphasis on questions about the admissibility of evidence on not the processes by which juries and judges move from evidence to findings of fact.
Facts are important in trials and regulatory proceedings, but they are also important in the lawmaking processes that occur in legislatures and in common law courts. Sometimes we call the facts that are found in trial courts "adjudicative facts" and the facts that are used in lawmaking "legislative facts." A similarly phrase is used to describe the facts that are used in the development of constitutional doctrines; these facts are sometimes called "constitutional facts."
So perhaps it is no surprise that most law students have never thought seriously about the logical structure of the factfinding process. My sense is that most students have a basic familiarity with the distinction between deductive and inductive arguments, but I would guess that many (or perhaps almost all) law students are unfamiliar with the idea of "interference to the best explanation" or "abduction." Indeed, at this point many readers might assume that this Lexicon entry is going to address an obscure topic that is only of interest to legal theorists: after all, how important could inference to the best explanation be, if law students, lawyers, and judges have never heard of it. But in fact, factual arguments based on inference to the best explanation are ubiquitous in the law. Knowing something about they way such arguments work can be eye opening and extremely useful.
As always, this entry in the Legal Theory Lexicon is aimed at law students, especially first year law students, with an interest in legal theory.
Deduction, Induction, and Abduction
The first step in understanding inference to the best explanation is to distinguish it from two other forms of argument, "deduction" and "induction." Almost every law student has a basic understanding of deductive arguments. A deductive argument is valid if the truth of its premises guarantee the truth of the conclusion; deductive arguments are sound if their logical structure is valid and their premises are true. Thus, "All dogs are mammals. All mammals are animals. Therefore, all dogs are animals." is both sound and valid.
Good inductive arguments do not guarantee the truth of the conclusion; instead, inductive arguments provide some degree of support for their conclusions. Here is an example:
- Every crow in a sample of one thousand crows is black. This supports the conclusion that all crows are black.
An entire field of legal study, "Empirical Legal Studies," is devoted to what we are calling "inductive arguments."
Unlike deductive arguments, arguments based on inference to the best explanation do not guarantee the truth of their conclusions. Unlike inductive arguments, inference to the best explanation does not depend on generalization from many instances (or samples). An argument from inference to the best explanation frequently moves from one fact (which might be an event or occurrence) to a conclusion on the basis that the conclusion is the best explanation for the fact that provides the premise for the argument. This may sound very abstract, but the idea is simple and intuitive in the context of specific examples.
Examples of Inference to the Best Explanation
The best way to understanding inference to the best explanation is via examples. Here are a few:
- Helberta wakes up in the morning and goes into the kitchen. The beans for coffee have been ground and placed in the filter bag in the coffee maker. Helberta is positive that they did not do this last night. The best explanation for this state of affairs is that Helberta's partner Gertruda set up the coffee the prior evening.
- Walter is found dead from a gunshot wound inflicted by a 45 calibre bullet. The day before, Graham and Walter had a violent argument during which Graham threatened to kill Walter. Graham owns a 45 calibre revolver and cannot account for his actions during the time period during which the shooting must have occurred. The best explanation for these circumstances is that Graham shot Walter.
- You are walking on the beach and see the following letters in sand: "L i f e i s b e a u t i f u l." It is is possible that the letters were caused by random wave action, but the best explanation is that another person drew the letters in the sand.
None of these examples involve inductive or deductive arguments. In each of the examples, it is logically possible that your conclusion is false. If you doubt this, consider the following additions to the examples:
- After Helberta sips her coffee, her sister Alberta comes into the room, surprising Helberta. Alberta says, "Sorry to scare you. I arrived late last night and Gertruda let me in. I got the coffee ready."
- After the police investigate Walter's shooting, they discover that Alice, Walter's wife had learned about the argument between Walter and Graham, had access to Graham's revolver, and was furious at Walter because they had learned that Walter had been sleeping with another woman.
- After you see the writing on beach, you come across a boy and their dog. The boy says, "Spot, write 'hello'!" and remarkably, the dog traces the letters "H e l l o" in the sand and these letters are similar in width and shape to the letters you saw earlier.
Arguments from an inference to the best explanation do not guarantee the truth of their conclusions, even when they make the conclusion seem very likely. Given new and different facts to be explained, the process of inference to the best explanation can lead to a different conclusion.
Inference to the Best Explanation in the Law
Inferences to the best explanation ubiquitous in the law. The idea of "circumstantial evidence" is based on inference to the best explanation. When someone says "the evidence is only circumstantial" they are point out the possibility that the inference to the best explanation does not guarantee the truth of the conclusion. But inference to the best explanation operates in the context of non-circumstantial evidence as well. Suppose that there is an eyewitness to a shooting (Helena) and that the Helena testifies at trial that Roberta was the shooter. The fact to be explained is Helena's testimony. If we believe her, it is because we believe that the best explanation for her having testified that Roberta was the shooter is that Roberta did in fact do the shooting, that Helena witnesses the shooting, and that Helena is now truthfully recounting what she saw. Of course, there are other explanations, and there is empirical evidence that eyewitness identifications of strangers are frequently wrong. More generally, inference from the testimony of witnesses to the facts about which they testify all depend on inferences to the best explanations.
The role of inference to the best explanation is not limited to factfinding at trial. The factual predicates for policy arguments may be based on inferences to the best explanation. And many doctrinal arguments about legal content use inference to the best explanation as well: the courts in cases X, Y, and Z, made decisions A, B, and C, and the best explanation for this patter of decisions is that the courts were implicitly following legal norm P.
Once you start looking for arguments from the best explanation in law, you will find them everywhere.
Inference from the Best Explanation and Behavior Economics
Just because inference to the best explanation is ubiquitous does not mean that it is unproblematic. Human psychology influences which inferences are seen as best, but the psychological mechanisms may be biased or flaw. Behavioral economics, which is grounded in psychology, identifies various heuristics that may result in inferences to the best explanation that are, in fact, invalid. For example, it may be the case that there is a tendency to prefer simple explanations over complex ones, but that the simplicity of an explanation is not a good predictor of its accuracy.
Conclusion
This Lexicon entry aims to introduce the idea of abduction or inference to the best explanation. The bibliography provides some references for additional reading and legal applications. There is much more to be said about this topic, but I hope that I have provided enough background to get you started.
Related Lexicon Entries
Bibliography
- James Hawthorne, Inductive Logic, Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/logic-inductive/ (2018).
- Igor Douven, Abduction, Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/abduction/ (2017).
- Harman, G., 1965. “The Inference to the Best Explanation,” Philosophical Review, 74: 88–95.
- Craig R. Callen, Spotting A Preponderance of the Evidence in the Wild: Inference to the Best Explanation and Sufficiency of the Evidence, 48 Seton Hall L. Rev. 1517 (2018).
- Ronald J. Allen & Alex Stein, Evidence, Probability, and the Burden of Proof, 55 Ariz. L. Rev. 557 (2013).
- Peter Lipton, Inference to the Best Explanation (2d ed. 2004).
(Last revised on August 4, 2024.)
