informal learning

Attack of the killer rhizomes

Oh my God, it's full of stars (93/365)

"Oh my God, it's full of stars"

Dave "the rhizome" Cormier posted the other day about uncertainty and rhizomatic learning. If you don't know what the latter is, Dave says it's

"A botanical metaphor, first posited by Deleuze and Guattari in A Thousand Plateaus(1987), may offer a more flexible conception of knowledge for the information age: therhizome. A rhizomatic plant has no center and no defined boundary; rather, it is made up of a number of semi-independent nodes, each of which is capable of growing and spreading on its own, bounded only by the limits of its habitat (Cormier 2008)."

In Dave's post he talked about habituation and how when you learn something it becomes automatic, and the vocabulary you adopt reflects this. In the comments I mentioned some of the work done with experts, in particular pattern recognition in chess experts. For instance, chess experts will be able to reproduce a board they are shown much better than you or I (assuming you're not a chess expert that is). The reason is that they encode it as patterns linked to long term memory eg "the mid-game position in Kasparov vs Karpov 1984" whereas novices are encoding it as discrete elements eg white rook next to black pawn two spaces in".

Experts don't know they do this, but it's a by-product, or rather a means, of expertise. They don't set out to encode in this manner, but it is what they do as they gain expertise. Interestingly, if you show expert chess players random placement of figures, ie not real mid-game positions, then they fare the same as everyone else.

Over twitter Dave said that maybe what he was talking about wasn't expertise in the traditional sense. To which I responded that in a very general, maybe even trite, sense, all learning is a shift away from novice towards expert, even if you never reach the expert level. No-one learns something to know less about it.

But then on reflection, maybe Dave has a point. The kind of learning he is talking about is accidental, acquisitional. It's almost collateral damage learning. I was thinking about being a blogger as an example. I didn't take a formal course to become one, obviously (I'd want my money back if I did). And I don't think I ever articulated to myself the goal or ambition to 'learn to be a blogger'. But over time, reading others, and experimentation, I learnt a lot of things, such as the right voice (for me) to use in a blog, what subjects to blog about, how to connect with others, the use of certain technologies, etc. I also became enculturated and learnt or adopted the 'blog culture'.

I suppose in one sense, I am now an expert, or at least more expert than some, blogger. But this feels rather different from other things I might be considered an expert in (erm, I can't think of any now, but there must be some), which have been the result of very intentional, and directional learning.

And this to me gets to the heart of the good and the bad about rhizomatic (or if you prefer, networked) learning. It works, at least in the sense that I have learnt to be a better blogger than I was six years ago. But if it's unintentional, undirectional, informal and accidental then is there much we as educators can say about it other than 'that's interesting'? My point is if we can't foster it, direct it, start it, measure it, or even look at it then is there much we can do about it? It may happen, and be very useful, but as soon as you try and touch it, then it disappears. It's like stars you can only see in the periphery of your vision, as soon as you focus on them, they fade from view.


  • dave cormier

    attack of the killer rhizomes indeed.
    Here’s what i’ve been thinking about the all important difference you’re describing. the two examples that seem to make the rounds are the chess master and the cab driver. Both of those ‘experts’ are involved in games that have rules and are bounded. They are bounded in many ways… but the critical one is ‘winning’. Chess is bounded by rules and has replicating patterns that have all been traced mapped and have their appropriate countermeasure. Same with the cab driver, traced, mapped and have their appropriate measure.
    Win by taking a king or delivering a passenger. Either way… everyone knows what success is.
    Is there more than one move that could win the chess game… sure… maybe, perhaps their are qualities of elegance or rapidity that are important.
    Is there more than one route to deliver the passenger efficiently. Perhaps safety or speed are relevant here.
    But in either case, winning is clear.
    In your blogging example, the rules aren’t clear. Winning could be just about anything. I’m comfortable with you calling it expertise… but you can’t be THE expert blogger. That’s the distinction i’m talking about.
    How many things in the world lack clear game rules? My suspicion is alot of them. They may be hard to teach for… but I don’t think it helps when we create artificial game rules, just so we can measure it.

  • mweller

    Oh sure – nearly all expertise is like that. Psychologists only study chess players because it fits neatly into an experiment, but their findings are transferrable to more typical domains. For example an expert in English Literature would encode a passage of text differently because they would be linking it to a body of work, style, etc. But one wouldn’t claim english lit is a game with tight rules.
    And of course, in almost any domain of expertise no-one person is THE expert. You can be an acknowledged expert in english lit without ever suggesting you are the expert, because such a notion is nonsensical.
    This much is still normal expertise and expert theory however, and nothing unusual. We can teach this stuff perfectly fine, and indeed have done for hundreds of years. My difficulty is that as I understand rhizomatic learning, it seems UN-everything – undirectional, unintentional, (erm) unformal, etc. So what do we teach here? It seems entirely resistant to notions of teaching. It happens, that seems to be all we can say about it?


    It’s not usual for me to disagree with Martin, so seeing this i had to bite. Here it’s more a case of ‘I agree with Dave’. Among knowledge workers, there’s a whole lot of learning going on where people are acquiring skills and expertise that makes them better at their job. They may find it difficult to recognise it as ‘learning’ (Eraut talks about this) but over time they change and become ‘expert’ and we can recognise that they have learnt. While I agree that it is difficult or impossible to measure reliably, and can’t be taught, I do think there is a lot which can be done to support ‘learners’ as they undertake this type of ‘learning’, and this is where we have tried to direct our recent work.
    In a knowledge intensive company we worked with, one of their grand challenges was to reduce ‘time to competence’ – the time between a new recruit joining the company and them starting to be of value (impacting the bottom line). Their new graduate induction programme actually did a pretty good job of addressing this challenge – not by attempting to teach the new recruits (the recruits came highly qualified) but by creating an environment which supported them in learning the culture, working practices and knowledge flows of the company. To do this they provided coaching arrangments for the new staff, programmed explicit opportunities for these new recruits to work in several different parts of the company in their first 12-18 months and ran a Graduate Network masquerading as a social hub through which recruits would develop new professional networks. The company recognises that these new recruits undergo a period of intense learning when they join the company and has created an environment to support them.
    We’re interested in the other things you can do to support these knowledge workers as they learn – what tools can you provide, what network structures can foster the social interactions that are so vital to their learning, can you help them balance their own learning needs with the need to deliver for the company.

  • mweller

    Thanks Colin – I wasn’t saying it couldn’t work this way, but it was more me asking a question and trying to work it through. If rhizomatic learning was so un-everything else, could we use it at all? Your answer seems to be that we can, if done correctly.

  • dave cormier

    I wouldn’t say it’s un-everything 🙂
    I would, for instance, say its very un-un-contextual. The context of a profession, being imitated as much as possible, will lower what Colin (quite nicely) calls time-to-competence. We need to become habituated to the language/technology/text/sign of a given domain in order to start the learning process. So at least un-un-context.


    There is more we can go than say it is “interesting”, we can consider it as a form of potential energy, what are they ways we can create learning environments that increases the potential for rhizomatic goodness?

  • Scott Johnson

    Hi Everyone,
    Late in on this. Seems that diffusion of knowledge or information is relevant here. Living organisms sustain themselves in a living state by taking on more and more strategies for survival. This runs against the trend for energy to drain from a system seeking ever lower states of excitement by shedding complexity. (See Autopoietic Networks).
    We seek information and also leak it back into the surrounding environment through our activities and this leaky vessel model seems as good a method of imagining rhizomatic learning as any. Boundaries and end states such as the achievement of expert status are less important here than competent, timely knowledge collected and applied where and when needed.
    And why can’t this be directed or thought of as a teaching method? In carpentry we used to sit around the job site with the apprentices and tell battle stories. Through this process the apprentices learned to model the role of expert and anticipate things beyond their experience that could both save time and the misery of accidents later on.
    This example maybe this is more suited to the definition of the effects of Legitimate Participation but it also suits my understanding of how rhizomatic learning allows that everyone is part way to being a competent practitioner of a set of knowledge long before the official designation of “expert” (permission to practice)is awarded.

Leave a Reply

Your email address will not be published. Required fields are marked *