Plato’s Meno begins with a question posed to Socrates by his eponymous interlocutor, asking whether virtue can be taught. Preparing for the last day of my advanced legal research course, I could not help but wonder the same thing with regard to legal research. I did of course teach the class (I certainly wouldn’t want my employer or supervisor to think that I was shirking my duties), but I do wonder what, exactly, my students learned from our discussions, assignments, and projects, and the feedback I offered. Through various assessments throughout the semester, they have demonstrated knowledge of a multitude of resources, they have planned and tracked their research, and they have thought critically about legal information – but do these things add up to being able to research the law, or is there something essential missing, something that cannot be taught?
Perhaps an appeal to authority can offer clarity about what may be missing. To borrow from a well-known treatise, “Legal research is the process of identifying and retrieving the law-related information necessary to support legal decision-making.”  A suitable definition – broad enough to encompass a range of information types and practice needs yet precise enough to accurately portray the role of research in legal practice – it hinges on the phrase “necessary to support.” This reveals that a (if not the) key part of legal research is exercising professional judgment to determine whether a particular bit of legal information is needed to decide a particular legal matter. Whether it is needed is a question of relevance.
Throughout the legal research process, the researcher is faced with multiple instances where he or she must determine whether a particular text, document, or provision is relevant to his or her research need. How does the researcher determine relevance in these moments? Relevance is an extraordinarily complicated concept. However, it should not be controversial to note that a researcher’s relevance determination, although contingent on a number of different factors (perhaps as many as 80 ), depends on his or her already existing knowledge of the particular subject or domain.
Those among us who teach first-year law students may be quite familiar with the importance of pre-existing subject matter knowledge (or its lack) in determining relevance. How can you teach legal research to students who know little to nothing about the substantive law or legal analysis and reasoning? They do not know what they need, they cannot recognize it when they see it, and they cannot fit it within a legal argument. This problem is lessened as students learn more about the law and its specific forms of analysis and reasoning, but it never really goes away. All research, after all, is rooted in some sort of information need, some lack of knowledge. How can a person recognize the relevant unless they already know what it is they are looking for?
This is a form of what is known as Meno’s paradox, or the learner’s paradox, appearing in the same Platonic dialogue referenced above. In Socrates’s formulation, the paradox reads as follows:
[A] man [sic] cannot try to discover either what he knows or what he does not know[.] He would not seek what he knows, for since he knows it there is no need of the inquiry, nor what he does not know, for in this case he does not even know what he is to look for. 
Socrates dismisses this as a “trick argument,” and others have poked holes in it as well. But the paradox does draw attention to the fact that learning is not a simple process that travels straight and smooth from seeking knowledge of x to finding knowledge of x.
Likewise, the legal researcher does not set out to research already knowing what it is he or she will find. Work must be done to articulate the information need, build up background knowledge, and place the problem within the larger framework of the subject domain. Legal research instructors can make this work easier by teaching students about secondary sources, research strategies, and the bibliographic and hierarchical structure of the law, but there is very little we as legal research instructors can tell them about how to determine the relevance of the items in a results list. Determining relevance is something the student must learn how to do on his or her own.
In light of the importance of determining relevance, legal research instructors should be addressing it in class, even if we cannot “teach” students how to do it. This may be as simple as offering an in-class acknowledgement of the paradox that limits both their attempts to determine relevance and our attempts to teach it. In the same vein, we can explicitly draw our students’ attention to the link between metacognition and relevance determination. The things that we can teach them – sources, skills, and strategies – are all learning tools that they can use to teach themselves the subject matter knowledge they need to determine relevance.
More than this, though, legal research instructors ought to begin thinking more about what constitutes relevant legal information and how its relevance is determined. Perhaps a generalized model of relevance determination for legal information can be constructed, a framework that can be used both to deepen our understanding and to serve as a teaching aid.  Maybe legal research cannot be taught, but we can always make it easier to learn.
 Steven M. Barkan, Barbara A. Bintliff, & Mary Whisner, Fundamentals of Legal Research 1 (10th ed. 205).
 Birger Hjorland, The Foundation of the Concept of Relevance, 61 J. Am. Soc’y for Info. Sci. & Tech. 217, 225 (2010).
 I’ve used the W.K.C. Guthrie translation of Meno 80e, published in Plato, The Collected Dialogues (Edith Hamilton & Huntington Cairns eds. 1961). A translation by W.R.M. Lamb can be found on the Perseus Digital Library.
 Marc van Opijnen & Cristiana Santos, On the Concept of Relevance in Legal Information Retrieval, 25 Artificial Intelligence L. 65 (2017) may be a helpful starting place.