Anemona Hartocollis in The NY Times asks, "Who needs education schools?" She states that education schools have research agendas rather than pedagogical, courses on theory rather than practical practicums, ideological biases rather than an emphasis on subject matter, and so on. One of the more interesting perspectives was that of a professor at Emporia State University in Kansas, one of the better schools:

Ms. Azwell says medical training is a good analogy for what Emporia State interns go through. "They get no sleep," she says. "They're working 24 hours a day. There are those who have no money because they can't work a job, so they're not eating. They're in a classroom 8 to 4 every day. They really think they're going to die."

I can appreciate the need for intensive training, but it's rather odd that getting no sleep and thinking you're going to die are characteristics of a program that will promote learning. Still, there is much to think about in this in-depth article, especially the general trend of education schools to focus on theory and ideology rather than giving future teachers the tools they need to be successful in the classroom. What are future teachers learning and how does that affect our schools?

In the article, Diane Ravitch, states,

The idea of "preparing excellent teachers who are excellent in their subject," she says, has been overtaken by other concerns - "professors wanting to be respected in the university, and teachers' colleges wanting to become places where research is done and to be agents of transformational change."

"At the end of the day, what would principals and parents value most?"

No doubt, Ms. Ravitch, has her own ideology, but it is worth considering what sort of balance should be achieved among research, ideology, and practice in schools of education.

  • What do we want to teach future teachers so that their future students will learn?
  • What sorts of academic practices lead to better learning?
  • Obviously, theory alone is not enough, albeit it is important. But should it outweigh actual practice in preparing teachers?

In Discover (via Kelly Creighton), Steven Johnson, co-founder and editor-in-chief of Feedcovers the positive effects of video games on learning. Essentially, they operate on the "competence principle"; that is, they bring learners to the higher edge of their competence and challenge them (rather than the lower edge and boring them). In addition, there's instant feedback on their performance. According to the article, video game players "see the world more clearly" and are "consistently more social, more confident, and more comfortable solving problems creatively."

This edge of competence reminds me of the edge of chaos, where complex systems arise out of chaos or restructure via a phase transition. And the notion of challenge fits in well with the motivational theories of Deci & Ryan and Lepper & Malone (see July 14, 2005 entry). Apparently, we're seeing a cross-domain phenomenon that complexity theory will be well suited for unifying. I wonder how that can further our understanding of learning theory and pedagogy.

I'm reading Alicia Juarrero's book Dynamics in action: intentional behavior as a complex system. One point she makes is that of "enabling constraints." Take the example of language in which rules, or constraints, on how sounds can be put together, enable meaning to be communicated. Without rules, language would be only noise. This perspective on learning is a good counterbalance to the prescriptive vs. "anything goes" dichotomy presented in pedagogy. The difficulty in a classroom activity, I imagine, would be determining ahead of time, what sorts of constraints would enable rather than disable.

Along with the germ cell concepts in the previous entry, we need to consider the flow of germ cell concepts in the classroom. Outside of the teacher, where are the key nodes in the network? Do they involve students? Bottlenecks could result from not including students in the network of knowledge flow in addition to not having appropriate germ cell concepts. So, the lever points somehow involve the interaction of students, ideas, and niches in the classroom. I'm not quite sure where to go with this. One point to consider is that in education today, there is a focus on the learner-centered classroom, as opposed to being teacher-centered. However, a better approach might be a network-structured, idea-focused classroom, or more simply, a learning-centered classroom. Along these lines, I recommend reading Engaging minds: Learning and teaching in a complex world by Brent Davis, Dennis Sumara, and Rebecca Luce-Kapler.

It may be difficult to see bottlenecks and lever points in the classroom, because one characteristic of classrooms is that all students learn and do the same things, leaving the teacher as the primary bottleneck/lever point. What if the classroom became more like an ecology in which different students occupied different niches? We see this sometimes when introducing technology and learning that one or two students are already proficient, we rely upon them to help the other students and us. But that remains a very non-diverse ecology. Project-based learning tends to have different students be responsible for different aspects of a project. One source of bottlenecks could occur in the flow of information between different parts of the project. But that seems to be a bottleneck for completing the project but perhaps not so much for learning. Bottlenecks, and simultaneously lever points, for learning would involve key concepts that are required before others can be learned, or at least would facilitate later learning. And so, we're back to building blocks and Davydov's germ cell concepts. What are the germ cell concepts for writing?

In Holland's model, tags direct the attention of agents toward certain features while disregarding others, thus facilitating selective interaction. In effect, tags identify and categorize phenomena, thus setting the boundaries of aggregates, or groups. Practically, that means that students recognize salient features of other students, thus forming groups that last the semester. Those features tend to be ethnicity, language, and gender. Tags are also values. When students work in groups, some value consensus models of interaction while others prefer more aggressive competition models. These values lead students to continue working with those of the same values and avoiding those of different values. With some groups, age tags operate. With the Chinese students I've had in my classes, the eldest one seemed to be a spokesperson for the rest, regardless of gender. Because tags regulate the formation of groups and networks, having an awareness of the mechanism of tagging can provide a new stance from which to see bottlenecks and lever points of classroom interaction and learning.

So, what/where are the bottlenecks and lever points of classroom interaction and learning?

On a listserv, we're discussing the concept of transfer in writing. We know that people learn and that they build on prior knowledge. But it's not clear in the field of composition how writing (whether skills or concepts) transfer to other classes and to careers. In fact, often the case seems to be that students do not transfer what they have learned in first year composition (FYC) to later classes.

Returning to the notion of building blocks and Davydov's germ cells, we can see that students need practice in adapting ideas from one context to another. But with only one semester in which to practice, which apparently is too short a time frame, I believe there needs to be a focus on those germ cells/building blocks that will be most fruitful in transferring, recombining, etc., along with practice in using them in a variety of contexts and genres.

Interestingly, it often takes only 3 or 4 types of agents (building blocks) to make a system quite complex. Consider the following: Four building blocks of DNA have led to the thousands of species on earth. All of physics can be explained by four forces: weak, strong, electromagnetic, and gravitational. There are four parts of speech: pronunciation, stress, rhythm, and intonation. Holland's model has 4 properties and 3 mechanisms.

Deci & Ryan, in their self-determination theory, says that motivation has 3 components: intrinsic motivation, extrinsic motivation, and amotivation. Extrinsic motivation is subdivided into a continuum of autonomy: integrated regulation, regulated regulation, introjected regulation, and external regulation--integrated regulation having the most autonomy and external regulation the least.

Learning task incorporating intrinsic motivation, according to Lepper & Malone (1987), include four elements: fantasy, control, curiosity, and challenge. Note that "control" overlaps with the autonomy (extrinsic motivation) in Deci & Ryan's theory. Hmm.

Those 2 theories are psychological theories of motivation. For a social theory of interaction (based on psychological mechanisms), Alan Fiske posits that there are four relational models: communal sharing, authority ranking, equality matching, and market pricing. These four models govern all social interactions, constrained by cultural guidelines. (There are also non-social interactions not covered by the theory.)

That so many different arenas can be based on a small number of building blocks suggests that when many factors are posited as explanations, it may be due to interacting combinations of 3 or 4 basic building blocks that generate the many factors.

Perhaps the lowest useful building block is words (or utterances) and then clauses (where does syntax fit in this model?). It's what comes after those levels that becomes more difficult to determine. Holland (Emergence, 1998) notes that poetry (and I include writing in general) has a looser framework than physics, which allows for a tight integration of building blocks. In Hidden Order, he gives the example of quarks, nucleons, atoms, molecules, organelles, cells, and so on, to show the integration of building blocks at different levels. Poetry's looseness, he says, "limits the possibilities for a cumulative structure," although not making it impossible.

For building blocks, I mentioned the questions of stasis theory below, perhaps in combination with Meier's Habits of Mind. Almost all academic texts would include these questions, or assume them in some fashion, and they seem to be natural ways in categorizing thinking, and so perhaps writing. If these are good building blocks, then we should see how various interactions/combinations of them can generate what we see at a higher level, if it exists. One emergent level should be genre.

Another potential contender, one I'm not familiar with, is Halliday's functional model with its concepts of ideational, interpersonal, and textual semantics. Gordon Wells (Dialogic Inquiry, 1999) interestingly tied it into activity theory's levels of operation, action, and activity.

Similar to building blocks are activity theorist Davydov's germ cell concepts in his "Ascending from the abstract to the concrete." This approach starts with students discovering primary general concepts in a particular discipline, investigating those concepts across particular contexts, and in the process retrace/recreate the process through which people developed the present day concepts.

I'd like to return to my first thoughts on Holland's model and why I'm interested in them. As a writing instructor I want to help my students improve their writing and be able to transfer what they learn in my class into other classes and eventually into their careers and elsewhere. Holland's mechanism of building blocks lend hope that we can achieve these purposes. That is, if we can determine (the) common building blocks of writing across a variety of contexts and genres, whether within school or without, and help our students master those blocks, so they can adapt them to fit in various combinations across new and diverse situations, then we have accomplished our goals.

Right now, I'm leaning toward stasis theory: What are the facts? What is the nature of the event? What is its value? and What should we do about it? And, I would add, how did it come to pass?

In a way, these questions are similar to those in Deborah Meier's Habits of Mind: How do we know what we know? Who's speaking? What causes what? How might things have been different? and Who cares? (or So what?)

The similarity between these sets of questions leads me to think that human thinking runs along a few fundamental paths (this is not new), and teaching writing along fundamental lines, i.e. building block, can facilitate our students learning these blocks and transferring them to new contexts, whether to other classes or to future careers. Thus, teaching composition will need to include both the content of building blocks and practice in adapting those building blocks to novel situations.

Charles Adamson with more comments (edited by me):

Charles Adamson , who has lived in Japan for one half of his life, suggests that Holland’s model is influenced by cultural and psychological factors. He writes:

Culturally, Holland, like most Americans and Westerners, generally divides the world into bipolar pairs, while people in Japan, for instance, frequently see continuums. Examples of Western bipolarities are animate/inanimate, sentient/nonsentient, good/evil, war/peace, etc. Psychologically ,based on Neuro Linguistic Programming (NLP), Holland processes information in terms of similarities, rather than differences as I do.

Point 1: It seems obvious to me that there are continuums from sentient to non-sentient and from animate to inanimate. However, for Holland, things are either this or that, with no middle ground. For CASs, he divides animate and inanimate, saying that only animate objects can be a CAS. He also classifies all animate objects together in terms of whether or not they are sentient.

Point 2: According to NLP, people tend to prefer one of two possible ways of processing information. They can look at the similarities, as Holland does, between people and bacterium, which lets him attribute human qualities (anticipation) to the bacteria. I process information by looking at differences, so I tend to compare bacteria to less animate objects and say, "Wait a minute. Neither of these are at all like a human, so you can not say that a bacterium can anticipate."

Points 1 and 2 together

My position and Holland's make sense as two different world views. One view is not more correct than the other, but one may be more useful in a particular situation. Holland's view leads him to his model of CASs, but mine seems to put limits on that model, restrictions that do not appear from Holland's viewpoint.

By limiting the use of the term 'anticipate' to only sentient beings, we can force a redefinition of the terms for CAS that include non-sentient agents, like bacteria. Once we have new definitions of the CAS properties and mechanisms that accept a non-sentient view of bacteria, I believe that we will find that a wider range of phenomena fall into the category of CAS, things like language for example.

So by a round about route, I have arrived at the position of Charles N a few days ago. I think that we need to modify, redefine the terms of, Holland's model of CAS, so that we no longer have the problem of words 'anticipating'.

My comments:

Certainly, different views may be more viable in particular contexts. So, I asked Charles, What would you like to redefine (in addition to “anticipate”)? Why? and What will be gained by doing so with respect to understanding CASs?

Before redefining, perhaps we should discuss what we mean by CAS: complex adaptive system (as opposed to complicated and merely complex systems). Can inanimate objects adapt? Rocks can't. What would we consider Brownian motion? It's a form of self-organizing, but it doesn't seem to be a CAS. Bacteria do adapt, so we need a CAS model that includes them. Words change over time, but do they adapt? Metaphorically (and what isn't a metaphor), it may be useful to see them as adapting. But then we need to consider what we mean by "adapt". Are we using the term with different meanings in different contexts? If yes, is it legitimate/profitable to do so? Does that mean that we are looking at different types of CAS? If yes, how is that different from Holland having different types of anticipation: implicit for bacteria and both implicit and explicit for human beings, perhaps with some continuum in between.

Charles Adamson responds:

This means that we have to be clear about what 'adapt' means and then to determine whether it applies to language seen as a time series. Maybe, it is not a CAS at all. Maybe, it is only the people involved that are CASs and the language is simply complex.

Charles Adamson later came up with a lengthy online definition of CAS from Kevin Dooley (saying that he would think some more now that he had a definition), a major member in the Society for Chaos Theory in Psychology and the Life Sciences. I've briefly summarized that definition.

"A CAS behaves/evolves according to three key principles: order is emergent as opposed to predetermined, the system's history is

irreversible, and the system's future is often unpredictable. The basic building blocks of the CAS are agents. Agents are semi-autonomous units that seek to maximize some measure of goodness, or fitness, by evolving over time. Agents scan their environment and develop schema representing interpretive and action rules. ... Existing schema can undergo three types of change. ... Schema define how a given agent interacts with other agents surrounding it. Actions between agents involve the exchange of information and/or resources."

My comments:

These schema are the same as Holland's internal models. For the most part, words undergo change, but perhaps it would be better to consider their tags as defining how they interact with other words rather than schema or internal models. So, with Charles Adamson, perhaps language is a complex system rather than a complex adaptive system. If so, then the primary differentiator seems to be the presence/absence of schema.

From the Chaosla listserv (a listserv dedicated to the study of chaos and complexity theory as applied to second language acquisition), Charles Adamson applied Holland’s model to language itself rather to the processes through which students produced it. What follows is our conversation with some paraphrasing, integrating of emails, and adapting for this forum.

Charles Adamson wrote:


Aggregation: This would seem to take place on a number of levels - letters into words, words into phrases, phrases into sentences, sentences into paragraphs, paragraphs into sections or chapters, and sections or chapters into complete works.

Nonlinearity: At every level language is nonlinear; the whole is not predictable for the parts. For example is we have the words the, man, dog, and kills. We can generate /The man kills the dog/ or /The dog kills the man/. We also have sentences like /The horse raced pace the burning barn fell/ which is almost impossible to understand the first time it is seen.

Flows: Initially, I was thinking that it was the flow of information, but maybe it is the control that each additional word in a sentence exerts on the potential words to follow. Each additional word in a sentence limits the pool of potential sequences that can follow it. Also it is obvious that the string of words has a strong role in selecting the tags that can be active for the following words. This might be considered the movement of resources.

Diversity: This refers to the variety of word and sentence types, parts of speech, etc.


Tagging: Words are tagged with both a meaning and a part of speech. These interact and determine the possibilities for the use of the word in context. I might mention that Robin Facett, a Hallidayan researcher, determined that there are something just over 300 slots in a generalized sentence. This means that there are just over 300 parts of speech since only certain words can go in each slot.

Internal models: This would seem to refer to the patterns that we can extract from the vicinity of a word. These patterns are strong enough that it is possible to generate an index number consisting of the sum of the inverse general frequencies of the three words on each side of the target word. This index separates the various senses of a word, in other words, the meanings.

Building blocks: Words and affixes, which become all the other things in language.

My comments:

Charles A.’s application modifies Holland’s model a little. In the model, internal models are mechanisms agents use to anticipate. Thus, if language is the system, we might consider words as "anticipating" (through tags) where they would fit in (or interact with) a particular aggregate of words. Although it doesn’t really make sense to me that words can anticipate. Even so, Brent Davis, a prominent complexity science researcher in mathematics education at the University of Alberta, considers ideas to be agents.

I’ll need to think some more about his suggestions concerning flows as those are concerned with the flow of resources among agents, but rephrasing him, it is an interesting idea to equate “enabling constraints” (another concept I acquired from Brent Davis) with resources.

Charles A. expanded more on internal models and the concept of anticipation:

The word 'the' will have an internal model where 'the' will be followed by modifiers (including a null modifier) and then a noun or nouns. This model will restrict the models of any following word to its internal model of being a modifier or a noun. Another example would be that verbs with their internal models that specify, among other things, the number of objects and whether or not the grammatical subject of the sentence is animate or inanimate.

I do have one problem with the use of 'anticipate' in relation to the internal models. Linguistically the internal model of the word 'anticipate' requires an animate, sentient grammatical subject. We can generally ignore this fact, but it is like proverbial rotten apple, given time it can cause all sorts of problems. It becomes very easy to start attributing other characteristics of sentient beings to the model. However, the model does not anticipate, it exists. We, the humans, anticipate when we think about the language processes associated with the word.

My comments:

Charles is right that problems occur when we apply attributes of sentience to inanimate agents. Holland draws upon biology for his model, and so, although it gives some insight to language, adaptations may be needed to use it with language as a system. I suppose we will need to see what is gained and what is lost when we do so.

I realized that it would be helpful to have Holland's model here on the blog rather than needing to refer to the papers, so I've put a slightly shorter version of it here.

Properties of complex adaptive systems

Aggregation has two meanings. One is simplifying complexity by grouping items with similar characteristics, a primary method, Holland points out, by which we build models. The second meaning refers to how complex systems behave. Through the combined interactions of less complex agents, complex collective behaviors emerge. Consider the following levels of aggregations, their interactions, and their emergent behaviors: cells, organs, individual human beings, and social and institutional groups.

The differences in behavior at different levels is due to the property of nonlinearity. In nonlinearity, the behavior of the whole cannot be reduced to the sum of the parts. Obviously, the behavior of individual human beings cannot be understood simply by studying cells and organs, and similarly the behavior of complex educational systems cannot be understood by the behaviors of individual members, whether students, teachers, or administrators.

A third property of complex systems is flows. Flows refer to the movements of resources among agents via connectors that vary according to the system. For instance, in a food transportation system, the connectors are the transporting vehicles; the resources are the different foods; and the agents are the transmitting, storing, and receiving entities, such as farmers and grocery stores. The elements in a network change over time as agents adapt to various situations. In a second language writing framework, flows include the movement of rhetorical knowledge among students and teachers within and across classroom boundaries.

The final property is diversity. Educational institutions consist of many different types of teachers (science, English, history), staff members (janitors, secretaries), administrators, and students. Diversity results from complex systems because each agent’s niche in the system “is defined by the interactions centering on that agent" (Holland). This diversity is a dynamic pattern because agents engage in progressive adaptations via their interactions with other agents, thus constantly changing their niches in the system.


The mechanism of tagging facilitates selective interactions and thus the formation of aggregates. Tags are identifiers and categorizers. They can be badges identifying people who work in a company, thus setting the boundaries of the aggregate. They can also be values that identify potential friends or mates for future interactions and screen out others. Tags, therefore, also influence flows because they almost always define the network by delimiting the critical interactions, the major connections.

A second mechanism is schemas, or what Holland calls internal models. Internal models are mechanisms for anticipating situations. Internal models develop from interactions with the environment through three steps: reproduction through fitness, recombination via cross-over, and replacement. If a schema is fit, that is, successful in anticipating situations and guiding behavior, it acts as a parent in reproducing new schemas. In recombination, parts of different parent schemas (i.e., building blocks, see below) cross over to each other and recombine to create, new offspring schemas. These offspring schemas can replace other schemas already in the population. If we translate these terms into composition classroom analogies, reproduction means that students continue to use schemas that work in their essays. Recombination refers to students incorporating, for example, new concepts of writing into their present concepts so that both old and new concepts are used in some hybrid form. Replacement refers to students replacing an older schema for writing with a new, usually hybrid, schema.

The final mechanism is building blocks. Holland gave the example of a human face, in which the common building blocks would include hair, forehead, eyebrows, eyes, and so on, up to ten blocks. He also broke each facial building block into ten alternatives (e.g., blue eyes, brown eyes, hazel eyes, etc.), which gave a total of 100 building blocks. If one were to choose one alternative building block from each of ten bags of facial components, there would be 10 billion distinct faces with only 100 building blocks.

Another example would be the four bases of DNA. Various permutations of these four building blocks have given rise to myriads of species, all uniquely adapting to and fitting their environmental niches. When applied to composition, Holland’s model of building blocks differs from those in which teachers “transmit” a static blueprint of writing to students who, in turn, learn to assemble various components in a linear, lock-step manner toward a predetermined product. Rather, in Holland’s model, the focus is on interactions, adaptation, and emergence. Like DNA, interactions between rhetorical building blocks and social environments generate species of arguments, each one adapting to social niches, such as political speeches, academic articles, newspaper editorials, and family squabbles. Thus, the interactions of a few building blocks can generate novelty and, as will be seen, learning.

As per the entry, below, it is the interactions of building blocks that I am looking at right now. Are there a few building blocks that can give theoretical rise to a coherent model of rhetoric for students in first-year composition, a model that they can take with them and transfer to new situations?

This entry is adapted from my article "Building blocks and learning."

In Holland's model, the mechanism building blocks are of particular interest to me as the concept suggests that most learning and creativity occurs recombining what is known rather than invention de novo.

The term building blocks may suggest a mechanical perspective on learning, but simply consider the myriads of living species that have emerged from various combinations of the four building blocks of DNA. And in writing, the repetition and reusability of building blocks, or patterns, allow for commonality across genres, while new circumstances fuel unique interactions between the patterns that generate novelty—and learning.

A major goal for me now is determining what building blocks in rhetoric are particularly fruitful for recombining. (This notion is similar to that of Davydov’s “ascending from the abstract to the concrete.”) Of course, students naturally select and combine building blocks on their own without direction from the teacher, and teachers present students with a variety of strategies and concepts to use in writing. However, a haphazard, cornucopia approach to pedagogy misses the point. Holland writes:

We gain a significant advantage when we can reduce the building blocks at one level to interactions and combinations of building blocks at a lower level: the laws at the higher level derive from the laws of the lower-level building blocks. This does not mean that the higher-level laws are easy to discover, any more than it is easy to discover theorems in geometry because one knows the axiom. It does add a tremendous interlocking strength to the scientific structure.

In other words, if there are kernel building blocks from which all other building blocks in composition can be derived, then learners, through a process of recombining them across novel and varied contexts, can gain a deeper, conceptual understanding of the discipline than they would otherwise. Contenders for building blocks might come from stasis theory, Toulmin logic, or the lines of argument of pathos, ethos, and logos.