Stephen Downes has been making a few waves lately with his recent brief summary of Connectivism as a Learning Theory, which is a great deal easier to read than his 616-page book on the subject (that I confess to only having skimmed, though I read and followed some of the articles that fed into it). David Wiley approves, but notes that the theory is incomplete. I approve too. Stephen is always a thought-provoking and interesting writer and this is his most succinct expression of the idea yet. But I'm not so sure that, as presented here, it is a learning theory at all. If it is - it is very hard to tell as it gets a bit fuzzy at precisely the point at which it seems to become one -  then it is one that appears either inconsistent or very likely wrong. This doesn't mean Connectivism is not a good and useful theory. I think it is, at least in most of the ways it has been presented by George Siemens, of which more forthwith. It's just hard to fathom it as a learning theory that makes any sense. So this is my attempt to make sense of what Stephen means and, at the end of it, to explain why connectivism (small 'c'), as George Siemens has explained it, is such a good idea not just despite but because it is not a coherent learning theory.

Connectionism

The Connectivist account of individual learning, in which the nervous system is understood as a neural network with emergent properties and behaviours resulting from its connections that we describe as 'learning', is certainly compelling. In fact, it is so compelling that it is accepted by most proponents of almost every theory of learning without blinking an eyelid and without any contradiction. We even have a word for it that has been around a great deal longer than Connectivism: connectionism.  There may be a few that believe in incorporeal souls or that, more plausibly, seek quantum explanations of consciousness but, even for these, a connectionist account is recognized as possibly incomplete but certainly true of how we think and learn at some level. This cannot be what Connectivism as a learning theory means or connectivists would simply call themselves connectionists.  Thus, though making a great deal out of the connectionist account to justify calling it a learning theory, Stephen asserts Connectivism's distinctiveness by extending that concept into our other networks, broadly lumped together as social networks. This too resonates strongly with existing theories, so I will attempt to describe how and why Stephen's Connectivism is a little different from these before discussing why I have problems with it as a learning theory.

Distributed cognition

If Connectivism is about saying that our individual intelligence or capabilities to function as social beings cannot meaningfully exist nor be meaningfully described without considering the people and objects with which we interact then it is again hard to disagree. We have a label for it: socially distributed cognition, a widely accepted and venerable family of models and theories that delves into the idea very deeply. This does not constitute a new or distinctive theory of learning either. More than that, if we are simply looking at first-order connections between individuals and the objects and people they interact with (the most basic hub and spoke model) then there seems no point in talking about networks at all in this context because none of the interesting things about neural networks have any meaning or relevance in a hub and spoke model. So that's not what Connectivism is about, though the fact that networks are not just composed of people, echoing Actor Network Theory as well as distributed cognition, is a significant part of the theory.

Collective intelligence

For Connectivism to make any sense as a distinctive learning theory, there must be learning in networks beyond our own first-order connections with them  - something important about the emergent behaviours of the networks themselves. This brings us into the very well-trodden field of collective intelligence, that looks at how the interactions of large groups of agents leads to emergence of behaviours and learning at a group/network level. There has been plentiful work on emergence in that context, as well as into how collectives learn. It is worth reading Howard Bloom's Global Brain, or pretty much anything by Scott E. Page for some in-depth examination of how that works, to mention just a couple of my favourites in a very long list indeed that stretches back to the early 20th Century and including work from many fields, including education. I mention these particular writings because they use very similar terminology and concepts to the six (I think - going on memory here) used by Stephen to characterize what makes social networks tick. Even I have made the odd contribution to this field (e.g. here). It's well worth studying but it is not something that Connectivism can claim as its own territory unless there is something more to it. That something appears to lie in its treatment of networks as a fundamental unifying principle. 

What makes Connectivism a distinctive learning theory

While the individual parts of Stephen's version of Connectivism seem not to be new and do not constitute a single theory of learning when viewed in isolation, that doesn't mean that there is no novelty in the way the parts are assembled. Indeed, that is how pretty much all invention happens. This then seems to be the crux of the issue: That Connectivism provides a unified model of how networks (including people's brains and their social networks) learn. This starts to look like the basis of a theory and seems more distinctive than any of the components so far. However, I think it is based on a spurious bit of reasoning and cannot ever work but, because it is a bit fuzzily portrayed, I may have misunderstood and apologize if I am misinterpreting things here. I would welcome clarification if I am wrong.

Topology is not equal to function

There are some topological similarities between brains and our social networks (including the mediating objects within them) but there are exactly the same kinds of topological similarity in the spread of disease, mob dynamics and the formation of traffic jams. There therefore has to be more substance to this idea than topological similarity. This is where things get sticky because, as Stephen is the first to admit, brains are different. However, he appears (this is the point at which it gets fuzzy for reasons I describe below, so I apologize if I misrepresent this) to wish to apply the same kind of principles that relate to neural networks, which have broadly uniform nodes, directed edges, constant distribution and qualitatively identical connections, invoking ideas that relate to neural networks like back-propagation, Hebbian rules and Boltzmann distributions as though they apply equally and similarly to the discontinuous, messy, asymmetrical, diverse, complicated world of social networks. His assertion seems to be that they are not exactly the same, but that they are part of the same class of explanations and, importantly, that learning happens within them in broadly related ways. But what does this actually mean? 

Brains are embodied neural networks of like nodes and directed edges

Brains have levels of emergent and structural organization that are tightly hooked into our bodies, with evolved intentionality to help us stay alive, look after children, eat, mate, seek comfort, avoid danger, learn to use tools etc. They have a purpose and that purpose is us (though, evolutionarily speaking, they may equally have a group-selection role too as we are a eusocial species). Technically speaking, they are directed networks. They are inherently contained, otherwise we die. Moreover, the things that make brains work are a specific kind of neural connection between the same types of entity. If a neuron could decide to behave differently from other neurons it would not be a good thing at all. Even a simple change in behaviour ('today I think I will reduce the strength of my signals' or 'I wonder what it would be like if I responded when things are quiet rather than when I get stimulation' or 'I'm going to talk back') would quickly degenerate into chaos and no thought at all if more than a few errant neurons began to diversify. Crucially, knowledge and learning in a neural network exists entirely within its configuration of connections, not in its individual neurons.

Social networks are diverse, plural, parallel and messy

Our social networks, including the mediating objects we create, are diverse, plural, parallel, reaching whatever emergent patterns they fall into by many different processes. They are discontinuous in time and space - this is not one distinct network but lots, of different kinds, different salience, different meanings. We move in and out of them constantly. We equally and constantly occupy multiple networks at once. We can choose when and whether to engage in them and are more often disconnected from them than we are connected. They can be directed and undirected, or mixtures of the two. They lack intentionality and purpose. Indeed, they often lack boundaries, save those that we choose to impose. Interactions between nodes are measured in myriad different ways and perform different roles, and nodes behave in many different ways and act as part of many different networks in dependent and independent ways, directed and undirected. Some of the nodes are intentional agents, with different agendas, and not all are nice. A malicious or misguided agent or two can really mess up a network.  Perhaps the most significant difference between social networks and neural networks then, from a learning perspective, is that in social networks knowledge exists in nodes whereas in neural networks knowledge exists in the network itself. To an extent that is also true of things like ants and termites that do exhibit complex collective behaviours while also having independent brains, but it is important to observe that the intelligence and learning capacity of an ant colony is not determined by the knowledge of individuals but by the algorithms for self-organization (e.g. stigmergy) that lead to emergent nest-level learning. It's the same kind of behaviour that leads to movements of money markets and share prices in human systems, which are equally smart and equally stupid for much the same reasons. Fascinating stuff and (as I and many others have shown) eminently exploitable in supporting human learning but not, I think, what Stephen means when he is talking about knowledge and learning in networks. If he is, then see my previous comments regarding collective intelligence. Suffice to say, the differences between social and neural networks go more than skin deep while their similarities lurk mainly on the surface.

Deep similarities, crucial differences

Of course there are underlying patterns and similarities between very dissimilar networks. Equally, we can usefully uncover and make use of general principles that apply to all networks of a similar topology. There is much to transfer such as patterns of self-organized criticality, stability and dynamism in small-world networks, principles of clustering, the roles of hubs and authorities and so on. This all helps us to make effective use of social networks for learning, to find strengths and limitations in them and to design or influence systems that make use of collectives to exhibit crowd wisdom in support of individual learning. However, though sharing some similar dynamics and topology, brains do something pretty cool that the spread of memes, the movements of pedestrians on sidewalks, the formation of ecosystems, the flocking of birds, the nest building of termites and social connections between people do not: they think. This is because they are utterly different networks organized in utterly different ways performing utterly different functions. To suggest they are similar is perfectly reasonable but it is no more or less relevant than saying that the fact that salt and sugar are similar because they are composed of electrons, protons and neutrons. In a great many important ways beyond this similarity they are alike, and it is indeed a little too easy to mistake one for the other, but you would not normally want to substitute one for the other in a recipe.

An apparent contradiction

For reasons that I think resemble the ones I have just given, Stephen goes to some lengths to disavow that notion that that social networks and neural networks operate in the same manner. But this is why it seems fuzzy to me because he also appears to be claiming fairly unequivocally that they do. After describing some network features as theories and stating that the physical make-up of the networks is immaterial, he claims that these are 'actual' learning theories and that learning is the formation of connections in a network. So, unless I have misunderstood him, he is suggesting that social networks learn in ways that can be explained or at least described by things like back-propagation, contiguity, etc, in much the same way as we describe neural networks.  So, as far as I can make out, Stephen is telling us that a social network is both not at all like a brain and very much like one. Such an apparent contradiction can only be true, without the Moon being made of green cheese, if and only if these claims relate to two epistemologically fundamentally different entities, which is exactly the problem that he has with other theories of learning. Unless, of course, he is talking about ways that social networks can exhibit collective intelligence, in which case I am fully on board (and wrote a book, a PhD and numerous papers about it) but that's another fundamentally different kind of entity, not directly about knowledge in a social network, and there are many other processes involved of which those relating to neural networks and their kin are but a very small if significant subset.  Therefore, either this is wrong, or this is not one theory but at a number of existing theories lumped together with only a common theme of networks to very loosely bind them or, as David Wiley suggests, it could be that it is just very incomplete. If so, it is much too incomplete to be described as  a learning theory, even if it does press-gang a bona fide learning theory (connectionism) into its service. I welcome correction if I am mistaken about this.

But it doesn't have to be a learning theory to make it a great idea.

A theory of how to learn, not a learning theory

I'm a big fan of connectivism (small 'c') in part because it is not a coherent theory of learning. Much more usefully, it is a situated set of principles, observations, perspectives and suggestions about how to learn, given the conditions that are made possible through the read-write web. It's thus a theory (using the term a little loosely but, I think, accurately) of how to learn, given a particular set of conditions, not a theory of learning.  This is an important distinction that is most visibly explicit in its constructionist values - you have to create and share stuff, not just because that's actually a good way to learn but, at least as importantly, because a learning network can have no value or content unless people actually share and create. It's how you do it, not what it is. Similarly for the cultivation of your network - it's a way of going about it, not a theory of learning. This is about how to use the network for learning, not learning itself.

Connectivism, as George Siemens formulated it, provides principles, models and techniques that, if applied, can help us to learn in a large-network context. George gave us a way of thinking about a related set of ideas that are relevant to structuring the learning process in a networked age. The process of learning in a connectivist account cannot be seen simply as something done in isolation nor just as something done through intentional group processes, but as a process of navigating and sense-making in a distributed complex adaptive system, in which that system, including its emergent as well as its designed properties, plays a first-class role in supporting, enabling and reifying learning (and the converse - mobs can be stupid as much as crowds can be wise). It is a context where more is different. George gave voice, shape and a name to a paradigm shift that was occurring and had been occurring for a decade or more before he started writing about it, including such things as communities of practice, distributed cognition, uses of complexity theory, heutagogy, constructionism, knowledge reification, knowledge gardening and much much more. My own PhD, started in 1997, was about very much this kind of thing and I was a very long way from being the first in the field (in fact I was quite peeved when George came up with such a good name for what we were doing because I had played with a lot of 'connect-' words in search of a broad defining term, finding all to be unoriginal, without hitting on 'connectivism'. Darn your brilliance, George!). Such notions were, in their turn, based on earlier visionary thinking from people like Bateson, Hofstadter and Illich, who lacked the adjacent possible of the Internet to make their ideas a reality. These ideas were in the air.

George gave us a single umbrella term that united a movement. He told the story with such clarity and vision that anyone could understand it. He gave those of us working in the area a banner to fly. We were doing something different from traditional social constructivist, instructivist, cognitivist and behaviourist teachers, different again from those working in the self-directed learning tradition, and it had a name. We were connectivists and George was our high priest. 

I see connectivism (small 'c') as a patchwork of loosely connected existing ideas, most of which are existing theories, most of which relate directly to learning, all of which imply a systems/complexity perspective and all of which contribute to a different way of understanding how to go about learning than models based on traditional intentional institutional or self-guided processes. There is nothing wrong with that at all. It is precisely what I and, I think, many others found so compelling about George's original formulation. Its fuzziness that linked a lot of disparate ideas together was a great strength, allowing many different and evolving approaches, theories, methods and processes to be seen as part of a unified movement.

The ideas connectivism patches together share a perspective, a way of seeing the complex ecosystem of learning in all its richness and, perhaps, allowing us to play a role in shaping how that ecosystem evolves. It is about networks, sure, but mainly in the sense that they are a prerequisite of a rich systems view of learning, a means of connecting with one another and the reifications of our knowledge, and a conduit for emergence, not because they are learning itself. Notably, connectivism gives us a perspective and an agenda that makes it easy to shift our gaze from formal education and/or an individualistic view of learning to one that almost demands different kinds of organization, taking inspiration from research into things like systems theory, city dynamics, ecological processes, evolutionary theory, complex adaptive systems and, yes, also brains. It's a perspective in which the individual learner not only matters and has active control, but contributes materially to the learning of others. Connectivism is thus incredibly important as a catalyst for change and offers great continuing value. But it is not yet close to being a viable theory of learning, nor do I think it is heading towards becoming one. In fact, I'd be a bit sad if it were, because then what would those who disagree on some of the important details call themselves?