#cck11: Language and logic

In ‘What Connectivism Is’, Stephen Downes says that connectivism is different from other theories of learning in not being ‘grounded in language and logic’. I think I understand what he means when he says the properties and constraints of linguistic structures are not the properties and constraints of connectivism. But it never occurred to me that they *would* be the same; I guess I see all theories of learning and knowledge as an attempt to put very complex ideas into words, and it’s extremely difficult to do this. One of the feelings I came away from this piece with is that connectivism, as a theory, is not quite as earth-shatteringly distinct from other theories of learning as Stephen makes out.

Tony Forster’s comment resonated with me: “I am not sure that Constructivism applies only to propositional learning, nor that all the symbol systems that we think with have linguistic or propositional characteristics.” Stephen’s response – that it would be “very difficult to draw out any coherent theory of constructivism that is not based on a system with linguistic or propositional characteristics”, is true – but doesn’t address the point Tony is making. How would *anyone* go about explaining something with no linguistic or propositional characteristics? Without language and logic, how do we communicate?

…Interpretative dance?

If Stephen is serious about this then perhaps he should offer up his argument in something other than the written form…? I would quite like to see that. Perhaps he’ll concede then that without a representational system or a syntax you don’t get a coherent *anything*.

I’m glad George and Stephen have joined forces to offer this course; they have different ideas and ways of explaining connectivism and, while I find George’s words easier to make sense of, I get plenty of fun out of grappling with Stephen’s. The notion that as soon as we make learning a deliberate process, it becomes harder to understand, is brilliant.

11 thoughts on “#cck11: Language and logic

  1. I’m inclined to agree with this – being new to learning theories I found basic Constructivist ideas almost immediately appealing and of quite general applicability whereas I can’t yet say the same for Connectivism. Your post has made me look and re-look at some of the philosophical debate surrounding connectivism and I can’t help feeling that it might be pushing a little too big for its boots. But early days, and I’m no expert. Maybe CCK11 will convince me otherwise – I’m sure an interpretative dance would help!

  2. Let’s not forget that interpretive dance is in itself a symbolic and representational system that is not universal 🙂 (I know you were kidding, but I thought I should throw that in).

    Having just finished a course in psycholinguistics last month, I am wondering if the linguistic representation was meant to be interpreted as a language-acquisition linguistic system and not a specific linguistic representation.

    I do agree with you that without some sort of system to represent what we know (or think we know) it’s hard to demonstrate to others who might need to know what we know 🙂

  3. “How would *anyone* go about explaining something with no linguistic or propositional characteristics?”

    This has been done many many times. Take a construct like a fractal. There is no linguistic unit of any type. Yet, you can write algorithms hat will reproduce the phenomena.

    Going to learning phenomena, you can check up the connectionist models of learning (learning in the cognitive domain rather than the instructional one) that connectivism is indirectly inspired from. Their chore hypothesis is that the knowledge is distributed within a network of heavily interconnected units. The knowledge was called sub-symbolic rather than symbolic because there is no one to one relationship between unit and symbolic construct. A node encode some part of a pattern. The pattern is revealed through a pattern of activation across a set of units. Yet, these sub-symbolic theories were described with words, precisely, with sufficient richness as to guide research work. Algorithms have been provided that let you model the process, in a reproducible way (given the same input and the same algorithm, you will get the same results). More info in this article (warning, academic) – no link allowed, go to google, search for: plaut “Connectionist models of development” filetype:pdf.

    As such Connectionism provided a theory of learning. With some limitations, though. It was possible to run a simulation and compare the results to data from experiments, it was quite difficult to describe what the model used for the simulation was precisely doing. The dominant criticism is that they can reproduce but not really explain (I grossly simplify). The problem there is the one of levels of analysis. This is nicely explained in a video on youtube. As no link are allowed, go to youtube and search for “brain based learning fad”/ These difficulties with providing an heuristic have to do with the fact that it was the first time that a new theory of learning (i.e., cognitive development) didn’t define a new theory of instructional design (i.e., learning in a classroom context). The notion of implicit learning, that came into the spotlight with the advent of connectionist learning never got much attention in the education domain. It explains why kids pick up a vast amount of knowledge in games, outside of any formal learning process.

    Connectivism doesn’t provide a theory of anything. Connectivism is the (new) knowledge that may or may not emerge as people connect (assumed over the internet).

    As Gordon said, without a more precise description, there is no applicability. Without guiding principles, you cannot decide of an implementation. What new knowledge to expect cannot really be predicted. Without any mean of making prediction or defining objectives, you cannot evaluate.

    Emergence. Yes, this is a phenomenon that has been proven to exist. But defining the theory in terms of the connections between individuals bring you right into the problem of levels of analysis. If you want to get an understanding of the human behavior, then focus on the connected person, not on the connections that exist between persons. This is what gamification and social gaming do. They provide principles and recipes that can be used to define better instructional practices.

    Triggers, needs and desires
    – People want to feel important
    – People want better stuff (inducing scarcity, available only when you unlock xxx)
    – People want to achieve things
    — easy unlocking = immediate results
    — don’t make it too hard to level up or attain new experiences
    — small, incremental steps that see immediate results
    – People want to build relationships and reciprocity
    – People want to feel good about themselves (gift)
    – People want to express themselves
    – People want competitive advantage
    (source: game mechanics applied to service design and engagement marketing – no link allowed)

    As I recently enrolled in another course by Siemens, Learning analytics, I have to regret that they seems to be blissfully unaware of the progresses made in that domain. Which can explain the proposition that “only fittest can survive CCK”. Unfortunately, they do very badly when it comes to engaging you and keeping you engaged. That’s one of the may wholes in their theory. I remember that at some point Siemens mentioned the huge number of drop out. If the courses are run in such a way that for 99% of participants, the connections persists more than a few days, then there is no value in connectivism. Connections exist only when individuals actively connect. “Connectivism” doesn’t happen when individuals don’t join in. You cannot have a theory that promote the value of connectivism without discussing principles of engagement.

    There is a lot more to be learned from reading about the dynamics and mechanics of social gaming than from connectivism as it has been formulated. Go to Gamasutra for great articles (no link allowed, go to google, search for: gamasutra social gaming)

    PS. Seriously, no link allowed?

  4. Typing long text in a small box is not ideal. Reediting is to be done most carefully ;-).

    “If the courses are run in such a way that for 99% of participants, the connections persists more than a few days…” Was really meant to read as “If the courses are run in such a way that for 99% of participants, the connections don’t persist for more than a few days…”

  5. There is another fundamental flaw in that article “What is connectivism”. Downes introduces it as a theory of learning. As educator, our focus is rather on instructional design. Our task is not to understand how learning happens, that’s what cognitive scientists do (cognitive psychologists, neuroscientists). Our task is to use the knowledge provided by others as to how learning happens to design instruction that works. Our job is to implement.

    If you take constructivism there is typically a distinction between constructivism, the paradigm that offers a perspective, or more exactly, view of the world and constructivist learning, the principles, recommendations and practice that guide the implementation of instruction.

    If you take constructivist, cognitive theorists like Piaget and Vygostki introduced a new paradigm (different from behaviorism) that posits that “Learning is an active, constructive process. The learner is an information constructor. People actively construct or create their own subjective representations of objective reality. New information is linked to prior knowledge, thus mental representations are subjective.”

    Constructivist Learning (which should really be treated as constructivist instruction), on the other hand is a theory about instructional design. How you should go about designing instruction given that you have this constructivist view of the world. It can be broken down into a set of dynamics and mechanics. Call them instructional strategies if that makes it any easier.

    – A teaching-strategy that views the learning process as one that considers experiences and contexts that make the student willing and able to learn → learning-readiness principle, differentiation principle, adapt practice.

    – The approach to learning is typically one of learning-through-action and often one of learning-through-collaboration
    – Elicit a learning strategy of Induction: Collect the irregular information and build principle concept.
    – Structure information so that it can be easily grasped by the student (spiral organization) → discovery principle, progressive-disclosure and scaffolded guidance practice.
    – Often relies on an Inductive-inquisitory approach to information presentation.
    – Design Instruction to facilitate extrapolation and or fill in the gaps (going beyond the information given) → learner-control, socratic-questioning practice, dynamics
    – Use raw data, primary sources, manipulatives, physical, and interactive materials → authentic
    Engage student in dialogue with the teacher and with each other. → learning-through-social-interaction
    – Use socratic-questioning techniques. Student autonomy and initiative are accepted and encouraged
    learning objectives
    – Students build and use their own knowledge. Real Situation in solving internal problems and investigation

    Learning activities
    – Simulation and induction tools

    – Elicit Higher-order thinking

    “Rich” environments
    – Most constructivist approaches advocate what Perkins (1991) terms “richer learning environments” (p. 19) as opposed to the “minimalist” classroom environment, which depends on the teacher, textbooks and prepared materials (Roblyer, Edwards & Havriluk, 1996). According to Perkins, most constructivist models use any combination of the following five basic resources.
    – Information banks – allow access to required information (e.g., books, encyclopedias)
    – Symbols pads – supports learners’ short-term memory (e.g., notebooks, laptops)
    – Construction kits – facilitates manipulation and building (e.g., Legos, Tinkertoys, Logo)
    – Phenomenaria – allows exploration (e.g., terrarium, computer simulation)
    – Task managers – gives help and feedback when tasks are completed (e.g., teachers, electronic tutors)

    As a guy who encourages educators to become socially connected, Downes has great value. To put a name on what he does facilitates adoption. That’s all there is to Connectivism, really. Let’s face it. Connectivism has strictly no value whatsoever as a theory of learning (that is, of knowledge acquisition) that could be used to guide research in the field of cognitive psychology. “Connectivism is, by contrast, ‘connectionist’.” Oh, great, apples are, well apple-like! “Knowledge is, on this theory, literally the set of connections formed by actions and experience. It may consist in part of linguistic structures, but it is not essentially based in linguistic structures, and the properties and constraints of linguistic structures are not the properties and constraints of connectivism. […] in connectivism, there is no real concept of transferring knowledge, making knowledge, or building knowledge” . Clearly, Downes is not proposing a new cognitive theory of learning. This passage suggests that he views knowledge as never built.

    In his role as an educator, Downes’ focus should be on providing a good theory of instructional design. As such, it is more beneficial to get the discussion going at the level of teaching strategies. What designers call mechanics and dynamics. Nothing prevents you from using a mechanic of “learning through social interaction” for some lessons and then a mechanic of “learning through problem solving” for other lessons. These are just tools in your toolset. Learn how best to use each one of them.

    • Hey Frances – thanks so much for the link – I found the bullet points on principles of connectivism *very juicy indeed* – the flavours of connectivism communicated in a concise summary!

      I’ll definitely keep a eye on IRRODL; it’s not a publication I’ve been particularly aware of before and I imagine there’ll be a few gems to be found in there. If you remember, do give me a shout when the special issue’s out!

      Thanks again 🙂

  6. Nice tautology, if it cant be explained in words it cant be explained.
    Seems every theory i ever heard of needed to be explained in words.
    Nonetheless, can it be experienced?
    I think it can.
    Cant really condemn something for not being a theory though unless theory is defined. Here it gets tricky.
    If definition of theory is predictive, then no its not.
    If theory is about rendering something less opaque, then maybe…
    But does it do anything not better explained elsewhere, thats the rub.

  7. Pingback: CCK11 Connection with networks and communities | Suifaijohnmak's Weblog

Leave a Reply

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