Theoretical Synergies
A reader asked:
Does connectivism contradicts AT?
No doubt different people have their own theories, but I have argued in the past that one of the major differences between connectivism and constructivist theories generally is that in connectivism learning is a property of the system, something that happens all the time, and is not therefore the subject of intentional activity. You don't decide to learn now, and maybe to not learn later, you are learning all the time, it's what the brain does, and the only choice you exert over the process is what you will do to affect the experiences leading to your learning. Watch TV all day and you'll learn about game shows and daytime dramas, practice medicine and you'll learn to be a doctor. Similarly, where constructivists say "you make meaning", I disagree with the expression, because the production (so-called) of meaning is organic, and not intentional.
Does connectivism aims to resolve these contradictions
There are of course no contradictions in nature. A contradiction is a linguistic artifact, the result of sentences believed to be true each entailing that the other is false. Because so much of cognition is non-linguistic, it is probably not useful to speak of contradiction in this context, but rather to speak of harmony and disruption. (I say this almost off-the-cuff, but this would really be a significant change in our understanding of logic and reason).
All connectionist systems - ie., all networks, as understood computationally - work though a process of 'settling' into a harmonious state. What counts as harmonious varies depending on the precise theory being implemented. For example:
- Hebbian associationist systems settle naturally into a state where neurons or entities with similar activation states become connected
- Back-propagation systems adjust according to feedback
- Boltzmann systems settle into a stable state as defined by thermodynamic principles
The 'disharmony caused by change' is best thought of as a new input that disrupts this settling process. The network responds to this change by reconfiguring the connections between entities as a result of this input. This is learning.
Whether we are able to address linguistic artifacts, such as contradiction, with a given learning experience, is open to question. There is no reason to expect a contradiction to be resolved, though were our linguistic artifacts based in experience, such a resolution would be a desirable, and expected, outcome.
Does connectivism need to be feed by constructs of other theories?
I think it's important to understand that connectivism is the adaptation of educational theory to these other theories, that it points to a theme underlying these other theories, and is not distinct from these theories.
Connectivism is, in my mind, a particular instance of a much broader theory of networks. Thus, evidence that informs us about the theory of networks generally also informs us about connectivism.
This is an important point. Constructive approaches to education (and most other things) place a special significance on the role of theory, and particularly the role that theory plays in providing a perspective or 'lens' through which a phenomenon is experienced. Hence we expect any given theory to provide a given 'stance', provide analytical 'tools', and beyond certain constraints (such as non-contradiction) no one theory is assumed to constitute a privileged stance. Theory-construction thus becomes an importance scientific and pedagogical activity, leading to a host of other constructs (such as, say, 'identity').
Connectivist learning is very different. It is not about creating cognitive constructs such as theories. Learning, according to connectivism, is a process of growth and development or networks rather than a process of acquisition and creation of concepts. Networks are not concepts. Concepts are represntational systems, they postulate a devide between what they are and what they represent, they therefor entail a theory of signs, or semiotics, and have linguistic properties (such as the law of non-contradiction). Networks are physical systems, not cognitive systems. Though they can be depicted as representing things (eg., a brain state may be thought of as representing a physical state), this depiction is in itself an interpretation, and not a property of the network itself.
Now I think that network theory in general and connectivism in particular can provide a set of tools to analyze *other* phenomena - I describe these as six elements of critical literacies, but the exact nature is unimportant here - but it is rather akin to the way mathematics offers us tools for the evaluation of other phenomena - mathematics can define data and instrumentation, such as measurement, ratio and comparison, and bookkeeping - but it would not be reasonable to turn these phenomena around as a means of evaluative mathematics.
Networks, in other words, are what they are. Network theory is nothing more or less than a description of networks, and the application of that description to other phenomena, just as qualitative theory is a description of properties (such as colour, size, shape, position, relation) and quantitative theory is a description of number and ratios.
I read an article that was published in March 2011 edition of IRRODL: “Proposing na Integral Research Framework for Connectivism: utilising Theoretical Synergies”, from B. Boitshwarelo (Botswana).
I founded it very interesting. However, certain questions have arisen in regard to the analysis done by the author.
1- Accordingly to Activity Theory (AT), learning is initiated by intention (p168): “learning as conscious processing, a transformational process that results from the reciprocal feedback between consciousness and activity”. Is this true in connectivism? Connectivism says that learning is the ability to construct and traverse networks. Sometimes, this process may not be intentional. I mean, sometimes we learn without being aware that this is happening (as a child, for example). Does connectivism contradicts AT?
Does connectivism contradicts AT?
No doubt different people have their own theories, but I have argued in the past that one of the major differences between connectivism and constructivist theories generally is that in connectivism learning is a property of the system, something that happens all the time, and is not therefore the subject of intentional activity. You don't decide to learn now, and maybe to not learn later, you are learning all the time, it's what the brain does, and the only choice you exert over the process is what you will do to affect the experiences leading to your learning. Watch TV all day and you'll learn about game shows and daytime dramas, practice medicine and you'll learn to be a doctor. Similarly, where constructivists say "you make meaning", I disagree with the expression, because the production (so-called) of meaning is organic, and not intentional.
2- (p 169) The author says that one feature of connectivism is that it recognizes the need to adapt to the ever-changing nature of information “in order to resolve the disharmony introduce by such change”. My point is: does connectivism talk about this? Does connectivism aims to resolve these contradictions or is about to accept and learn to live with them? Are connectivist systems stable?
Does connectivism aims to resolve these contradictions
There are of course no contradictions in nature. A contradiction is a linguistic artifact, the result of sentences believed to be true each entailing that the other is false. Because so much of cognition is non-linguistic, it is probably not useful to speak of contradiction in this context, but rather to speak of harmony and disruption. (I say this almost off-the-cuff, but this would really be a significant change in our understanding of logic and reason).
All connectionist systems - ie., all networks, as understood computationally - work though a process of 'settling' into a harmonious state. What counts as harmonious varies depending on the precise theory being implemented. For example:
- Hebbian associationist systems settle naturally into a state where neurons or entities with similar activation states become connected
- Back-propagation systems adjust according to feedback
- Boltzmann systems settle into a stable state as defined by thermodynamic principles
The 'disharmony caused by change' is best thought of as a new input that disrupts this settling process. The network responds to this change by reconfiguring the connections between entities as a result of this input. This is learning.
Whether we are able to address linguistic artifacts, such as contradiction, with a given learning experience, is open to question. There is no reason to expect a contradiction to be resolved, though were our linguistic artifacts based in experience, such a resolution would be a desirable, and expected, outcome.
3- (pp.171-172) Is it really necessary to use the theoretical concepts of other more consensual and tested theories to study and validate connectivism? Does connectivism have his own tools of analysis to do this? Does connectivism need to be feed by constructs of other theories? Doesn’t this contradicts connectivism as new approach to learning in the digital age?
Does connectivism need to be feed by constructs of other theories?
I think it's important to understand that connectivism is the adaptation of educational theory to these other theories, that it points to a theme underlying these other theories, and is not distinct from these theories.
Connectivism is, in my mind, a particular instance of a much broader theory of networks. Thus, evidence that informs us about the theory of networks generally also informs us about connectivism.
This is an important point. Constructive approaches to education (and most other things) place a special significance on the role of theory, and particularly the role that theory plays in providing a perspective or 'lens' through which a phenomenon is experienced. Hence we expect any given theory to provide a given 'stance', provide analytical 'tools', and beyond certain constraints (such as non-contradiction) no one theory is assumed to constitute a privileged stance. Theory-construction thus becomes an importance scientific and pedagogical activity, leading to a host of other constructs (such as, say, 'identity').
Connectivist learning is very different. It is not about creating cognitive constructs such as theories. Learning, according to connectivism, is a process of growth and development or networks rather than a process of acquisition and creation of concepts. Networks are not concepts. Concepts are represntational systems, they postulate a devide between what they are and what they represent, they therefor entail a theory of signs, or semiotics, and have linguistic properties (such as the law of non-contradiction). Networks are physical systems, not cognitive systems. Though they can be depicted as representing things (eg., a brain state may be thought of as representing a physical state), this depiction is in itself an interpretation, and not a property of the network itself.
Now I think that network theory in general and connectivism in particular can provide a set of tools to analyze *other* phenomena - I describe these as six elements of critical literacies, but the exact nature is unimportant here - but it is rather akin to the way mathematics offers us tools for the evaluation of other phenomena - mathematics can define data and instrumentation, such as measurement, ratio and comparison, and bookkeeping - but it would not be reasonable to turn these phenomena around as a means of evaluative mathematics.
Networks, in other words, are what they are. Network theory is nothing more or less than a description of networks, and the application of that description to other phenomena, just as qualitative theory is a description of properties (such as colour, size, shape, position, relation) and quantitative theory is a description of number and ratios.
Stephen, thanks for the fine clarification of the role of networks in thinking about education, knowledge, and so forth. You are correct to point out that Connectivism is a subset of the broader discussion about networks; however, I fail to follow your thinking on a couple of points.
ReplyDeleteFirst, you say that there are "no contradictions in nature. A contradiction is a linguistic artifact," which might suggest to some a divide between nature and it's constructs and human constructs. I'm uncomfortable with this distinction, though I know it is the dominant view. I see it useful to consider linguistic structures as a part of nature: a complex layer of the total network emerging from a physical substrate, and perhaps having different rules than that substrate, but still dependent on that substrate. It seems to me best to think of language in all its variations as a part of the natural mix, part of what the Universe has created. Isn't it more useful to think that language and linguistic structures—even contradictions—have naturally emerged in nature, and are therefore natural?
This leads to the second statement that I don't quite follow: "Networks are physical systems, not cognitive systems." But aren't cognitive systems also usefully thought of in terms of networks? And aren't cognitive systems based on physical substrates, physical networks? Isn't it more useful to think of cognitive structures as just a different scale of the natural network? I agree here with Olaf Sporns that "cognition is a network phenomenon."
Again, I think I'm having trouble with what I sense in your statements as a separation between the physical and natural and the cognitive. Is this distinction necessary? Or have I missed the point of your argument? Thanks.
> I'm uncomfortable with this distinction, though I know it is the dominant view. I see it useful to consider linguistic structures as a part of nature
ReplyDeleteThat's very much a minority view, and I think not one you can maintain consistently (or course, that may not be a problem for you).
The thing is, the contradiction is not a part of the physicality of the expression. It arises only as a result of the interpretation we place on the symbol system, as a result of how we apply truth, meaning, and other abstract properties to the expression.
In the physical world itself (at least, according to the way we use words as they normally mean) it is not possible for something to be both P and not P. It can't be both a dog and not a dog. Yes, you can alter your linguistic system to allow contradiction - that's what your stance does - but you cannot successfully incorporate that stance into the physical world. That's a fallacy I call "the linguistic pull" - the belief that physical systems are governed by non-physical laws.
> It seems to me best to think of language in all its variations as a part of the natural mix, part of what the Universe has created.
But you recognize that it would be absurd to say "there are pink dragons," right? because someone imagined them to exist, believed them to exist, or uttered the statement "pink dragons exist."
The multiplicity of linguistic systems does not entail a multiplicity of physical systems.
> But aren't cognitive systems also usefully thought of in terms of networks?
This is a variation of what Dennett would call 'the intentional stance'. But it is also an example of what Paul Churchland would call 'folk psychology'.
Here's the dilemma:
If we use the word 'dog' in a relatively ordinary and unambiguous way, we can realtively easily create a mapping such that statements about 'dogs' are statements about a definable set of physical objects [dogs], such that what is true of 'dogs' is also true of [dogs].
For the intentional stance, or folk psychology, to be successful, then we need also to be also to do the same thing with common abstract concepts.
The problem is made most clear with a concept like belief. We can all use the word 'beliefs' and have some sense of what it means. But there is no set of objects [beliefs] such that we can map from 'beliefs' to [beliefs].
The same sort of problem exists with logico-linguistic terms such as 'truth' and 'meaning'. Again, there is no mapping from 'truth' to [truth] (the best attempt is Tarski's theory, which would be best represented here as "'snow is white' is true iff [snow is white] is 'true'.")
We know that there are human brains, and that they do things like think and evaluate and intend. But the terminology we use to describe what human brains is terribly imprecise. That's fine, so long as we don't do what you do here - so long as we don't infer from properties of the symbol system to properties of the physical system. The 'physical symbol system' hypothesis, in other words, is false.
So while cognition is indeed a network phenomenon, it is not governed by the principles and rules we have to this point characterized as cognitive phenomena.
>In the physical world itself (at least, according to the way we use words as they normally mean) it is not possible for something to be both P and not P
ReplyDeleteI sense a weakness in your argument in this statement....
It's an interesting post, thanks. If connectivism is the adaptation of something Educational, shouldn't there be more of a focus on maximizing learning and intentional learning, rather than just learning itself?
ReplyDeleteThis seems to me a very big gap in connectivism. We can't simply connect, we have to connect in some way...be it through language or other representational systems. Although these systems are surely not the same thing as what they represent, they play an inseparable part in what learning will take place. Connectivism may be a description of potential learning, but the quality of actual learning needs to consider how we connect.
> We can't simply connect, we have to connect in some way...be it through language or other representational systems.
ReplyDeleteIf I bonk you in the head with a thrown apple, we've connected - even though no language or representational system was used.
This is what's important about connectivism (and network approaches generally) - the connection itself, rather than any putative 'content' of that connection, is what's important.
I guess the place to start might be with an examination of what types of connections exist. A classification of connections, or description of them. There seem to be at minimum two: representational, and physical...
ReplyDeleteIf you bonk me in the head with an apple, you've connected by throwing an apple at me. That's much different than connecting by throwing a wrench, or connecting by making a phone call.
ReplyDeleteWhen you say "the connection itself", do you mean to say "connecting itself"? As I read most of it, you're not concerned with the connection itself.
I would say both are potentially equally important in connectivism (pipe and content), because it is applied to a field. Throwing a wrench at me, compared to a nerf football is going to change whatever your intended message is in connecting with me, regardless of where that message originates.
> That's much different than connecting by throwing a wrench
ReplyDeleteYes it is. But it does not follow that there is a representational difference, or that the difference constitutes a representation.
> "the connection itself", do you mean to say "connecting itself"?
No. Connecting is the act of forming a connection. A connection is the result of the act of connecting. But There are minor semantic differences ('connecting' is a success verb) that I don't want to mix in with what I'm saying here.
> Throwing a wrench at me, compared to a nerf football is going to change whatever your intended message is in connecting with me
This assumes there is an intended 'message', ie., some content in the connection. But it does not prove that there is content in the connection.
Hmm … perhaps we are talking past each other. When I say that I see it useful to consider linguistic structures as a part of nature, I am saying that linguistic structures are built on, or emerge from, physical structures, and I fully recognize that the physical structures came first (I'm speaking in evolutionary time here). I also accept that physical structures and linguistic structures have different rules. I also accept that it's often advantageous for the linguistic structures to map as precisely as possible to the physical structures, especially if we're talking about physical structures.
ReplyDeleteI am, then, viewing linguistic structures as emergent from various physical substrates: neuronal patterns, vocal sounds, organized marks on stones, clay tablets, papyrus sheets, and computer screens. The principle of emergence is still contested in science, but it is not uncommon. In his article Emergent Biological Principles and the Computational Properties of the Universe, Paul Davies defines it as "the appearance of new properties that arise when a system exceeds a certain level of size or complexity, properties that are absent from the constituents of the system." My point is that linguistic structures, and consciousness in general, absolutely emerge from and depend upon physicality (or nature), and yet they also have properties that do not belong to the physical substrate from which they emerged. As you point out, linguistic constructs can contradict one another, whereas physical constructs cannot.
This orientation means that I would not likely make a couple of the statements that you make. For instance, you say that "contradiction is not a part of the physicality of the expression." I say that it is part and parcel of the physicality of the expression. I do not know how to form a contradiction without a physical expression. I am contradicting you now, but only as I'm typing these electro-mechanical symbols and as you are reading them. Of course, I contradicted you earlier (yesterday, in fact) as I was thinking through your comments, but even that contradiction absolutely depended upon my physical neuronal structures, among other things (I also think the coffee I had should be factored into this equation, but I won't pursue that just now). Thus, while contradiction is a property of linguistic structures and not a property of its physical substrates, contradiction cannot exist without that physical substrate.
Sorry, but I exceeded the limits of the comment box. I end this way:
ReplyDeleteI am, of course, sympathetic to the distinction you make between the physicality of an expression and the abstract meaning of that expression when you say that, for instance, contradiction "arises only as a result of the interpretation we place on the symbol system, as a result of how we apply truth, meaning, and other abstract properties to the expression." However, I'll contradict you again. I don't think that meaning is some independent, abstract entity that "we place on the symbol system." Rather, meaning is what emerges as we spark networks of physical neurons and build networks of physical symbols and then move those symbols through larger networks to connect with and affect others. Thus, abstract meaning can do things that physical words cannot. Thus, I do not intend to "infer from properties of the symbol system to properties of the physical system." I do believe in what Davies calls strong emergence, in which "higher levels of complexity possess genuine causal powers that are absent from the constituent parts. That is, wholes may exhibit properties and principles that cannot be reduced, even in principle, to the cumulative effect of the properties and laws of the components." Thus, language can both do things that can't be done at the physical level and it can cause things to happen at the physical level that might not have happened otherwise, but it is still an emergent feature of the physical level and part of that level. I don't see how abstract meaning can exist without the physical, natural world. Though I can certainly imagine such a thing.
And this brings me to the last contradiction I'll make. You say that "it would be absurd to say 'there are pink dragons.'" Well, yes, but only in the narrow sense that you probably used the term absurd to indicate a cognitive construct that does not map with any rigor, regularity, or reliability to any physical construct. Yet, if I allow only language that passes your test of the absurd, then I eliminate much of both poetic and rhetoric, the twin pillars of my professional and personal interests. Fact is: pink dragons do exist in literature. Well, perhaps not pink ones, but certainly the common, everyday, brown or gray type of dragon—St. George's and Tolkein's. To my mind, it's absurd to say that dragons do not exist in imaginative literature, and it's equally absurd to say that imaginative literature does not exist in nature and is not absolutely dependent upon physicality.
Moving imaginative literature—or imagination in general—into the realm of the natural allows me to apply network principles to my study of imaginative literature. So far, I have found that very useful and productive.
@Stephen Throwing the apple represents your intention to get my attention. If you're doing it just for fun, then it represents your idea of humor. If, like you say, you do it for no reason then whatever it represents or not (geometric shape of the universe?)is beside the point here because it's not intentional...which Education is. This is the particular instance.
ReplyDeleteI think I get what you mean about the "connecting" definition. Although, I still find it confusing as it seems to include actual connections, not just potential ones.
I think that Education has an intention of promoting connectivity for the purposes of knowledge transfer.
ReplyDelete> throwing the apple represents your intention to get my attention
ReplyDeleteNo. This is just speculation on your part. I might just be practicing my aim. And that's my point. The attribution of 'a reason' or some 'meaning' to my action is something *you* are bringing to it, not something that was inherent in the act.
@Keith, I am happy to say that meaning is an emergent property of the communication or communicative act.
ReplyDeleteI've thought a lot about emergence over the years.
I think that one of the key things about emergence is that for us to say some phenomenon is emergent we have also to say that it is *recognixed* as such.
For example, the Jesus face on Mars is an emergent property of light and rock outcrops. But it becomes a 'Jesus face' only if we already know about Jesus. Otherwise, it's just random light and dark.
In other words, an emergent property is not inherent in the system producing it, but depends entirely on the perceiver being able to recognize it. (And this leads directly to a definition of knowledge - to 'know' is to be able to 'recognize').
"depends entirely on the perceiver being able to recognize it"
ReplyDeleterecognize the attribtion that where added ?
a random pattern of dots or a more face like pattern to a HD image of a famos jesus statue , are both in the media and in the mind.
the post remind me of the 2003 argument with Mark H. Bickhard about representation (that is how i found this blog).
the reason there are no contradiction in nature is a diction a statment (by proven theories).
however contested difrent theories on are, its agreed that only the most settled are realistic in the sense that there is a thing such as nature.