Investigating MOOCs, Part 2 of a #CLMOOC List

This is the second in a series of posts about Pierre Dillenbourg’s 1999 article What do you mean by ‘collaborative learning’?, which introduces his book Collaborative-learning: Cognitive and Computational Approaches. I became interested in Dillenbourg when a Rhizo14 Facebook conversation favorably referenced him and this article. I wanted to see if his approach to studying learning provides guidance for those of us who are exploring rhizomatic learning, and I was delighted when Dillenbourg opened his article with a discussion of scales within learning groups, as I consider cMOOCs a function of complex, multi-scale networks (actually, I consider most everything a function of complex, multi-scale networks). Any investigation of a cMOOC must be aware of the scale or scales from which it is considering the cMOOC and must accept that its focus on any particular scale likely obscures or distorts equally relevant and valid processes, qualities, and affects/effects at work on other scales and that its own scale likely interacts in significant ways with those other processes, qualities, and affects/effects. Dillenbourg’s article, then, provided me with at least one positive approach to studying cMOOCs.

However, in her comment to my first post, Jenny Mackness points out that Dillenbourg is talking about collaborative learning and from her point of view most cMOOCs are all about cooperative learning. Jenny writes:

Cooperation in the sense of open sharing was more what the original cMOOCs were about. And indeed Stephen Downes at the time (2008) made a point of drawing attention to the difference between collaboration (groups) and cooperation (networks) and warned against the dangers of groups and group-think. … For me - MOOCs are still about cooperation. I don't participate in a mOOC for collaboration. For me collaboration might (and often has) been the result of my cooperative experience within a cMOOC.

She is exactly right. cMOOCs are mostly about cooperative learning, in the sense that cooperation tends to emerge in the cMOOCs I engage and that cooperation keeps me engaged; thus, we should rightly question if Dillenbourg’s article about collaborative learning is relevant. I think his observation about scales is relevant, though his subsequent observations may not be. Still, they are worth reviewing as they may provide a useful contrast to other approaches to online rhizomatic learning. I have to keep in mind that this particular article is from 1999, 5 years before the emergence of Web 2.0 and almost 10 years before the first MOOC. At the time, Dillenbourg likely had no frame of reference for online, cooperative, self-organizing learning groups numbering in the thousands from across the world. So I want to look at the second and third elements of Dillenbourg’s collaborative learning, keeping in mind that he almost certainly did not anticipate cMOOCs.

The variety of meanings for ‘learning’: Dillenbourg says that collaborative learning is usually defined as either a pedagogical method or a psychological process. He defines collaborative learning as neither, or rather as a bit of both:

the words ‘collaborative learning’ describe a situation [italics in original] in which particular forms of interaction among people are expected to occur, which would trigger learning mechanisms, but there is no guarantee that the expected interactions will actually occur.

The key to collaborative learning, Dillenbourg continues, “is to develop ways to increase the probability that some types of interaction occur”, and he suggest four ways to encourage interaction:

1. to set up initial conditions: controlling for group size, group members, group arrangement, groupware used, suitable tasks, and so on.
2. to over-specify the ‘collaboration’ contract with a scenario based on roles: forcing students to play different roles in some activity such as a discussion.
3. to scaffold productive interactions by encompassing interaction rules in the medium: supplying structured responses (“I propose to …”) for students to complete or a set number of blog posts and comments.
4. to monitor and regulate the interactions: performing minimal pedagogical intervention to redirect student work in a productive direction or to encourage students left out of the interaction. 

As I understand him, then, the situation is a pedagogical method, and the learning mechanisms are the psychological processes triggered, or not, by the pedagogical methods. To my mind, Dillenbourg is not defining learning. I know his main topic is collaborative learning, but his sub-title is the variety of meanings for ‘learning’. I expect a definition of learning, but all I see is that learning, whether alone or in a group, is a psychological mechanism of some kind sparked by some given situation that encourages interaction among people. To be fair, Dillenbourg does note that we learn not because we are alone or in groups, but because we “perform some activities which trigger specific learning mechanisms” such as induction, deduction, compilation, explanation, disagreement, mutual regulation, and so forth. Still, he says very little about what this learning is. This does not encourage me. I’m discouraged even more when he characterizes interactions as social contracts between student peers or between students and teachers. I sense that I will have to work hard to realize some benefit from Dillenbourg’s thoughts here. I am almost certain that I will have to push beyond the way he unpacks these ideas.

The way Dillenbourg characterizes learning in this article suggests to me staged collaboration in a closed class rather than open cooperation in the sense of a cMOOC or even open collaboration in the sense of Wikipedia, an xMOOC, or the Rhizo14 auto-ethnography. He says that “basically, collaborative learning takes the form of instructions to subjects (e.g. “You have to work together”), a physical setting (e.g. “Team mates work on the same table”) and other institutional constraints [italics in original] (e.g. “Each group member will receive the mark given the group project”).” I don’t know that this definition holds today even for collaborative learning in a classroom, but I’m certain it does not hold for online cooperative learning. This concept of collaborative learning is based on the traditional notion of a prepared course which marches a cohort of students (or subjects) through a fixed course syllabus toward some specified knowledge or skill and delivered, managed, and assessed by an authoritative teacher under the auspices of a sanctioning institution. The instructions to subjects (students) come from the teacher as an authoritative representative of the sponsoring institution. The physical setting, even when aided and mediated by computational devices, assumes a proximate cohort bounded by space and time. The institutional constraints embed the entire process in an institutional framework. This is all so brick-and-mortar.

This is NOT a criticism. Rather, it is 1999. And in the short 15 years since, cMOOCs have pushed education beyond what our brightest minds were able to conceive and frame. Cooperative, online learning is basically the opposite of the collaborative learning that Dillenbourg describes.

First, cooperative, online learning does not take the form of instructions to subjects. As the recent Rhizo14 and DS106 have demonstrated, open cMOOCs can function quite nicely without an authority figure giving instructions to his subjects. (Yes, I'm playing with the regal overtones of the term subject here, which I don't think Dillenbourg implies, but it makes my point. My apologies if I offend.) Open, online learning is certainly full of suggestions, often too many to process, but these suggestions must be framed quite differently from the instructions in a traditional educational system.

Then, cooperative, online learning does not have a physical setting, or rather is not limited to a physical setting, a prescribed space and time. It is not even limited to an online setting. Space and time become very fluid concepts in open, online learning systems. Participants engage as they can with the tools at hand, when they can, where they can, with whomever they can. This relaxation of a specific and specified space and time is quite disorienting to many when they first engage an open, online space. This fluidity disrupts the normal relationships among people, who become confused about who is in and who is out (I think the whole freaking world is in, but that's just me), perplexed about how to relate, when to assert, when to hold back. The familiar social contracts seem to melt, and attempts to reform them seem to flounder. Not only do the familiar relationships among people fade, but the relationships to content become disjointed. Am I learning? If so, what? (As Bob Dylan once said, "There's something going on here, but you don't know what it is, do you, Mr. Jones?").

Finally, cooperative, online learning has almost no institutional constraints: no grades, no certificates, no ceremony, no sanctioned value. If you want value out of a cMOOC, then you have to create it for yourself with others. No one tells you the answer.

What are the implications here for research into cMOOCs? Mostly, we have to rethink everything. First, we must rethink the relationships among the participants. The old social contract assumed stable, discrete entities who formed stable relationships made explicit by contracts. Any violation in the terms of the contract dissolves the relationship. In the cMOOCs I like, leaders and groups wax and wane, emerge and fade. I constantly shift from lurker to speaker to leader to follower to in the group to out of the group, as do others. Relationships are very fluid. Are we learning together, or are you researching me? Sometimes it's hard to tell, so how do I relate, and do explicit rules make much sense? Finally, we must rethink how we determine the value gained from a course with no answer and no one purpose. How do I know when I'm finished? Am I still in Rhizo14? Have I really started CLMOOC?

It's hard to say, but that's what we are proposing to study.