Several of the disagreements with my recent post were with the (apparently low) threshold I set for learning. I continue to believe that if (1) I don’t know something, and (2) after engaging in some process I do know it, that (3) I have learned something. It doesn’t matter to me how insignificant the thing I now know may appear to someone else. It doesn’t need to be the fundamental theorem of calculus or a complex skill like driving a car to qualify as learning. Learning a person’s name is learning. Gaining the ability to successfully complete any kind of paired-associate task, whether matching a name to a face, or a capital to a state, or a country name with a position on a map, is learning – even if you can describe it using pejorative terms like “memorization.”
If you’ve ever tried to learn the names of the 50 students in your class, you know that this takes a significant amount of study and effort. If learning one name doesn’t qualify as learning, then simply repeating that task 50 times can’t qualify as learning, either. If we deny simple paired-associate recall the exalted status of Learning, the entire edifice of more complex learning is sacrificed on the altar of the sorites paradox. Because self-organization can’t emerge ex nihilo: “this globally coherent pattern appears from the local interaction of the elements that make up the system.” If there are no elements, whither emergence? In other words,
If there are no nodes in the network there can never be connections in the network.
The entire edifice of “complex learning” is built on the foundation of “simple learning.” To me, the nodes in the network are the “low level” kinds of learning (aka memorization) Ebbinghaus was studying as he established the forgetting curve, or that Anderson recognized one hundred years later when he saw reflections of the environment in memory.
You really can’t argue with their empirical work; you can only try to deny their subject status as Learning. But I don’t see a need to do that. Without memorizing basic vocabulary, you can’t develop a more complex language skill like conversation, etc. If you believe knowledge is like a graph, then the only way to increase the theoretical connectivity of your graph is to increase the number of nodes. If the important, sexy learning that we really care about is all connections and relationships, isn’t the first step in that process increasing the number of nodes in the network?
Any learning, whether expert-guided or self-directed, that increases the connectivity of the graph we call knowledge without simultaneously increasing the number of nodes in the graph is anemic.
Everything old is new again; maybe Bloom was on to something.