It’s Never Too Late to Learn about Learning Theory
But Why Didn’t I…?
I’ve learned more about theories of learning two weeks into an excellent four-week summer course than I did in all of graduate school.
Why didn’t anyone tell me this stuff was so good?
Why University Professors Don’t Learn about Learning
As Cathy N. Davidson and Christina Katopodis observe in The New College Classroom (2022), “most people who become college teachers have not received much training as teachers. … It’s rare even to find an academic outside of education departments who’s aware of the extensive learning science research. … Graduate students, including those preparing for a lifelong career as instructors, are rarely taught or exposed to pedagogy at all” (15). Ellen Schrecker documents the historical origins of this problem in passing in her book The Lost Promise: American Universities in the 1960s (2021): during the (so-called) “Golden Age” of Higher Education (the 1950s-1960s), Humanities faculty at leading universities consciously decided to emphasize research and minimize or altogether abandon teaching. They didn’t just choose the German Research Model over the earlier Liberal Arts model of training clergy (the usual story), they also fused elite study of the Humanities with the R1 system of resources and rewards.
Not all “R1s” are resource-rich, of course. “R1” is my perhaps somewhat-lazy shorthand for resource-rich institutions, where you’d have faculty salaries starting at around 85-100k+ for Assistant Professors in English, 2/2 or lower teaching loads, generous faculty-development funds, junior sabbaticals, etc. This R1 research (as opposed to teaching, often even anti-teaching) aspirational model then dispersed throughout most all elite graduate training—and its mostly elite programs that supply the faculty ranks of most universities. In “Where Do PhDs in English Get Jobs?” economists David Colander and Daisy Zhuo show the prestige networks that drive hiring models throughout the discipline, from R1s to lower-ranked regional colleges.
So material history, resources, and perceived prestige combine to interpolate (read: assume a set of values as natural norms) graduate students in the Humanities into ways of life that minimize teaching or assume it as a kind of natural ability. Elite liberal arts colleges can throw this model off slightly since they emphasize a teacher-scholar approach but with resources more akin to elite R1s.
Nils Gilman rightly identifies the fusion of personal identity and the research function that this model produced:
Elite universities in particular have built their faculties almost entirely around research achievement, with teaching treated as a secondary obligation. Reconceiving the professoriate will mean altering tenure criteria and promotion incentives, and it will face fierce resistance from scholars whose professional identities are bound up in the research function.
I’ve experienced this “fierce resistance” firsthand at major conferences when I’ve gently pointed out the possibility that the field’s largely elite-determined incentive structures can cause a neglect of teaching in favor of research. There’s ample evidence for this: Think of how people still say you can “Write your way out of” one lower-ranked university into a higher one, but no one ever says you can “Teach your way out of” one. Think of publishing in PMLA or NLH (research) versus publishing in Profession or Pedagogy (teaching)—Profession’s website is still on Winter 2024-5, PMLA is current. The research/teaching divide doesn’t have to be an either/or choice. But I’ve found that many (not all) of those in the elite model can treat appeals for a greater focus on teaching as a false dilemma where such a shift would mean neglecting research.
History has bitten back pretty badly since effective teaching often requires near-constant legitimation (“Why are we doing this?”), often for audiences without a buy-in incentive (i.e. a grade in a graduate seminar or possible recommendation that predisposes them to assume the Why?). Legitimation is one thing the Humanities seem to need desperately right now (here, here, here, here, and here). But the R1 prestige and rewards model largely creates internally-facing, research-for-research’s sake legitimations. “Public Humanities” names the desire for research-oriented faculty to have their work (i.e. their research) face outward to a wider audience, but it often tragically ignores the “public” right in front of them: their undergraduates. Instead, it implicitly blames a “public” for not valuing research.
Once you realize all this, it can be productively humbling to admit you, as a college professor, might not have thought much about your teaching at the level of how learning theory shapes teaching practice.
But turning to learning theory in the undergraduate setting will both (1) make us better teachers and (2) concomitantly afford us better legitimation for why what we do is important to others (or force us to realize that some of what we do isn’t or is no longer important—or is only important to us).
Behaviorists in Humanist’s Clothing
As a result of not receiving training in learning theory, a lot of professors, I suspect, often those at smaller liberal arts colleges or universities, think of themselves as humanists, both in the philosophical and (perhaps unbeknownst to them) in the learning theory sense. But, at least from a learning theory perspective, I wonder if they can often actually function as behaviorists who think they are humanists (with a touch of guided constructivism). Shaped by Watson and Skinner, for classical behaviorists, “observable behavior, not internal mental processes or emotional feelings, determines whether learning has occurred” (Bierma et al., 2014, p. 26). For humanists, “human beings have the potential for growth and development … and … are free to make choices and determine their behavior” (Bierma et a.l, 2014, p. 29). Shaped by Maslow and Rogers, in humanist approaches to learning, “the focus is on the inner person, that person’s needs, desires, and wants and how these require attending to in any learning environment” (Bierma et al., 2014, p. 29). Shaped by Dewey and Vygotsky, constructivists see learning as “the construction of meaning from experience” (Bierma et al., 2014, p. 36). Starting with each of these would produce a different learning activity, although, when composing learning objectives, you can easily see how the different approaches can appear in, and also shape, different parts, including the classical approach to Learning Objectives, Audience, Behavior, Conditions, and Degree of Mastery.
We can stretch “observable behavior” to include students reproducing the “correct” opinions and positions when they complete their writings and contribute to in-class discussion. A philosophy class or sequence that wants students to think that classical philosophers are virtuous heroes while moderns are undermining villains is, it seems to me, behaviorism in humanist clothes. So is an English class that wants students to believe that Alexander Pope’s writings are the correct sensibility while Kurt Vonnegut’s are postmodern popular fare. Education in sensibility can run the risk of being behaviorism that thinks its humanism. Or maybe it’s more charitable to say that such approaches are humanist behaviorism constructivism, using the tools of behaviorism to inculcate or suggest the superiority of some variation on humanistic philosophy as the “right” or “best” philosophy for life all through the process of constructivism. But I suspect this can often reduce to simple behaviorism, or as Bierma et al. put it for behaviorism, “arrange environment to elicit desired response” (2014, [p. 39). If the desired response is “Students come to believe idea A,” that strikes me as classical behaviorism. This doesn’t mean that this codes as “bad”; it just means that it approaches learning as content transmission rather than discovery or pattern-recognition or creating meaning from experience.
Connectivism and Rhizomatic Learning
My own preference and pedagogy leans more to a recently-emerged learning theory called connectivism, which, according to Christopher Pappas, developed out of the work of George Siemens and Stephen Downes in the 2000s and responded to “the enormous growth of information and the increased dependence on technology,” which moved knowledge from “a static process of simply delivering content” to “a dynamic activity that involves numerous digital and social interactions” (Pappas, 2025, n.p.). As Pappas describes it, Siemens and Downes saw learning as “a process of building connections across multiple sources of human and digital information,” where “the focus of learning [shifts] from the individual to the collective … moving from internal recognition to external connections, illustrating a paradigm in which learning happens across a broad spectrum of people, online communities, databases, and devices” (Pappas, 2025, n.p.). Pappas encapsulates the promise of connectivist approaches to learning; they require “metacognitive strategies for processing knowledge” and they are “closely aligned with how people learn and work in real-world scenarios” (Pappas, 2025, n.p.). Connectivism thus fuses humanist conceptions about truth and value with practical applications and (again) Jim Carroll’s “knowledge velocity” behind the wheel of change (2022, as cited in Bierma et al., 2025, p. 32). But, in connectivism, the “truth and value” part of humanism seems largely reduced to assessing the accuracy and validity of information –it’s not necessarily a higher meaning or morality or teleology of the kind Pope Francis might imagine.
At first, connectivism seems to have its, ahem, roots in the rhizomatic method of Gilles Deleuze and Felix Guattari where, like a root network but also not at all like a root network, there are infinitely multiplying connections: “A rhizome ceaselessly establishes connections between semiotic chains, organizations of power, and circumstances relative to the arts, sciences and social struggles” (Deleuze and Guattari, 1972, p.7). Unlike what appears to me to be the “linear” thinking of syllogistic reasoning in scholasticism, which proceeds from a to b to c, etc. and often reaches its desired conclusions, rhizomatic thinking can plug into existing networks at any possible point and begin creating connections. It’s my own preferred teaching method because it leads us to questions like “How does A relate to X relate to Y relate to 10, etc.?” It requires the work of mediation, what Adorno famously told Benjamin he needed. Works of literature are almost always plugged consciously or unconsciously into vast cultural, historical, political, economic, religious, etc. networks, and the work of interpretation can make these networks visible—and find new ones.
But rhizomatic learning seems to be distinct from connectivism. In a blog post from 2015, Siemens attempted to distinguish the two by saying that (1) both were organic, but, in admitting he hadn’t read, but was going to read, Deleuze and Guattari (2) critiqued rhizomes as a kind of “monoversity,” as opposed to diversity. But that’s definitely not what Deleuze and Guattari imagine them as being—exactly the opposite: “The rhizome itself assumes very diverse forms …” (Deleuze and Guattari, 1972, p.7).
A more germane distinction might be that Deleuze and Guattari’s work was primarily diagnostic and part of the general “postmodern” task of critique, which sought to productively undermine and subvert reigning notions of “natural” hierarchy and value. Connectivism instead seems to take knowledge as already consolidated into stable but shifting points (nodes) that are consistently reforming into new sets of links and thus new networks. It seems driven by globalization and digitization, almost making a learning virtue of capitalist necessity. As Siemens writes, “Everything – from fixing my car, to my morning coffee, to my research, to my mobile phone, to healthcare – is a function of connected specialization. Novelty and innovation arises when we collide ideas or specialties that previously had not been brought in relation to one another.” Siemens’s examples are maybe a bit telling in that many are commodities or costly items or procedures. The goal is to stay current; always be keeping up, etc.
As Alaa A. lDahdouh et al. summarize connectivism,
In connectivism, the structure of knowledge is described as a network. The network is a set of nodes connected to each other. These relationships/connections may not be seen as a singular link between two nodes. Instead, they are more like patterns: groups of relationships that come together as a single whole. The network is not static; it is dynamic and those patterns may change over time. Learning, according to Connectivism, is a continuous process of network exploration and pattern finding; it is a process of pattern recognition. (2015, p. 14)
Think of knowledge as “nodes and links” that can be brought into larger constellations through myriad connections.
As lDahdouh et al. make clear, connectivism has run into a problem with how to think about knowledge creation, not simply “pattern recognition.”
It might be that less anxiety about rhizomatic learning or drawing on the values of humanism or at least an anti-technocratic ethos can help untether a promising learning theory like connectivism from capitalism’s logics.