Two Kinds of Knowledge
Posted on Hiffington Post, November 19, 2010.
Why is the Republican Party now represented by red, when conservative parties in all other places -- and even the United States, in the past -- were represented with blue? Ben Zimmer suggests (via The Language Log) "Democrats may have wanted to appropriate the positive connotations of blue (as in true-blue)" but I wonder whether it isn't deeper than that. Because I recall over the years studies saying that teams that wear red win more frequently. Perhaps Republicans have deliberately chosen red in order to generate a subconscious association of themselves as winners.
That's speculation, but the association between political advertising and psychological preference is not. A recent Fast Company post describes the use of what they call "political neuromarketing" during the campaign. It's not really neuro marketing as it has nothing to do with neural connections. Rather, they "measure everything including the story line, level of the language, images, music. Using critical point analysis, [they] identify specifics that may drive voters away or attract them. The techniques are non-invasive, and include measuring muscle, skin and pupil response."
The success of such techniques obviously has its implications in political theory, but is also relevant in learning theory. The general principle that "the brain reveals more than spoken answers to questions" tells us that knowledge, beliefs, and other mental states are much more fine-grained than our more traditional analyses suggest. Understanding that learning -- and persuasion -- is not simply "words in -- words out" is the first step toward developing a more comprehensive theory of cognition and a more effective understanding of learning and instruction.
A recent paper from a group of leading neuroscientists outlines the understanding of learning beginning to take form. The survey paper brings together the results of dozens of studies of learning and cognition. The authors write, "Neuroscientists are beginning to understand the brain mechanisms underlying learning and how shared brain systems for perception and action support social learning." In some cases, this understanding is very detailed, such as our understanding of the function of layers of neurons in the visual cortext. In other cases, our understanding is beginning to cover a broad range of psychological phenomena, such as those involved in language learning.
It is tempting to use the analogy of a computer in an effort to understand human learning. That's why we see sentences like "the brain is a machine with limited resources for processing the enormous quantity of information received by the senses." But we should not even be talking about learning in such terms. As neuro-linguists will tell you "the brain does not store precise memories in specific locations. Instead, the brain reaches decisions through the dynamic interaction of diverse areas operating in functional neural circuits." The way we store, process, and represent information in the mind is completely different from the way it is done in a computer.
This is important because it tells us that learning is not simply, or even primarily, a process of decoding linguistic expressions. We can arrive at reasonable sounding generalizations about reading as decoding -- that we need to know that letters represent sounds, say, or that words have meanings -- but these generalizations do not lead us toward an understanding of language learning, they lead us away from it, as they are based on the supposition that cognition consists of word-like and meaning-like structures, which are applied to sounds and symbols, and refer to states of affairs in the world. But this just isn't so.
What we are in fact responding to as learners, especially at a young age, are patterns of perception presented to us from the environment. Children use frequency distributions, covariation and transitional probabilities to associate spoken words with phenomena. Learning, especially in the young, is imitative rather than analytical. Goals and objectives are inferred from patterns of related phenomena, not a propositional awareness of another's mental state. Phenomena are not experienced and understood in isolation, but in context and mediated by environment, social interaction, and previous experience.
It's a bit of an overgeneralization, but we can get at many of the issues here by distinguishing between two kinds of knowledge: one that is personal, internal to ourselves, and is, shall we say, 'knowledge-in-the-brain', and the other that is public or social, external to ourselves, and is, shall we say, 'knowledge-in-the-world'. Of course there are more than just two kinds of knowledge, but that is a discussion that can wait until later. The point here is to establish that there is more than one type of knowledge; if we can establish that, the rest can follow.
The distinction of these two types of knowledge refers to the nature of the knowledge itself, not the reality that the knowledge (putatively) describes. It is tempting to say that what we have here are two distinct representational systems, and if that works for you that's fine. But I believe the knowledge itself is the representational system, and so if we have two distinct representational systems, we have two kinds of knowledge. But let's not bog down on issues of ontology and metaphysics.
These two types of knowledge are well-established in science and philosophy. One of the more well-recognized versions of this distinction is articulated by Michael Polanyi (and echoed by knowledge management specialists everywhere). Public, social or external knowledge is what we might call 'explicit' knowledge, while Polanyi called the personal or internal type of knowledge "tacit" knowledge. This distinction has been characterized several ways. One way is to describe tacit knowledge as 'knowing how' while explicit knowledge is 'knowing that'. Another way is to distinguish between knowledge we can express and knowledge we cannot express. Tacit knowledge, argues Polanyi, is ineffable. It cannot be described. "We can know more than we can tell."
What's important about this distinction is that it creates a pretty clear dividing line between what we learn and what we express. The one is very different from the other. Expressions of knowledge are essentially the production of social artifacts -- what some would call "stigmergy" -- in order to coordinate activities with other people in the world. The results of this coordination constitute the rules of grammar, the laws of nature, etc., "the patterns of categories contain, theories, methods, feelings, values and skills which can be used in a fashion that the tradition judges are valid." These are phenomena, which can be learned, but the knowledge they express is expressed externally to the self.
What we learn, even when we learn from texts and documents, is distinct from the knowledge expressed in the texts themselves. Polanyi writes, "when I receive Information by reading a letter and when I ponder the message of the letter I am subsidiarily aware not only of its text, but also of all the past occasions by which I have come to understand the words of the text, and the whole range of this subsidiary awareness is presented focally in terms of the message. This message or meaning on which attention is now focused is not something tangible; it is the conception evoked by the text." The text says one thing, but when we read, we think of (and learn about) whatever is (in ourselves) evoked by the text.
When we learn, we do not merely assimilate; we do not simply undertake a mechanical process of decoding meaning from printed or spoken text. "Our knowledge of the things denoted by words will have been largely acquired by experience in the same way as animals come to know things, while the words will have acquired their meaning by previously designating such experience, either when uttered by others in our presence or when used by ourselves." This knowledge is not merely subsymbolic, it is distinct from the knowledge contained in the symbols. A doctor's knowledge of medicine is distinct from his or her knowledge of the words describing medicine. "While the correct use of medical terms cannot be achieved in itself, without the knowledge of medicine a great deal of medicine can be remembered even after on has forgotten the use of medical terms."
Tacit knowledge is learned using the visual cortex, cerebral cortex, and the rest of the neural network that constitutes our brain and nervous system. Knowledge, seen from this perspective, is not words and sentences or even pictures and icons, but sets of connections, layered over and over on each other, a fine mesh, a deep tapestry incredibly richer and more complex than any abstraction such as spoken language could express. As Nonaka and von Krogh summarize, "tacit knowledge is acquired with little or no direct instruction, it is procedural, and above all, practically useful." And while "locked away in people's neural networks," tacit knowledge expresses itself in our actions, our responses, and our expressions.
As Ryle said, "[T]o believe that the ice is thin is to be unhesitant in telling oneself and others that it is thin, in acquiescing in other people's assertions to that effect, in objecting to statements to the contrary, in drawing consequences from the original proposition and so forth. But it is also to be prone to skate warily, to shudder, to dwell in imagination on possible disasters, and to warn other skaters. It is not only a propensity to make certain theoretical moves, but to make certain executive and imaginative moves, as well as to have certain feelings."
This is an ability we share with animals, including some (like some primates) who can learn primitive languages, and others, like birds and cats (who cannot). And as Jeffrey Klugman recently wrote in Time, animals can learn a wide range of things once thought unique to humans. We've known for some time that animals can use tools, and have evidence of vocabulary and language in primates. But animals can also plan, work cooperatively, count numbers, have emotions, have empathy for others, and have a sense of self. And while humans may have specialized mechaisms for some functions (such as Broca's area for language) the mechanisms that produce this knowledge are low-level; for example, in problem-solving, "While the specialized cells in each section of mammalian basal ganglia do equally specialized work, the undifferentiated ones in birds' brains multitask, doing all those jobs at once."
Two different types of knowledge. Two different sets of skills. If we want people to socialize, to conform, to follow rules, we'll focus on the repetition of the symbols and codes that constitute explicit knowledge, to have them become expert in what Wittgenstein called "language games," the public performance of language. But if we want people to learn, then we need to focus on the subsymbolic, the concepts, skills, procedures and other bits of tacit knowledge that underlie, and give rise to, the social conventions. We cannot simply learn the words. "A great deal of medicine can be remembered even after one has forgotten the use of medical terms."
Or, to put the same point more bluntly, we can teach to support learning, or we can teach to support the production of social artifacts. We can teach the subject, or we can teach superficial behaviours. And as Tom Hoffman notes, those who have deeper knowledge, a greater base of expertise, will tend to produce "deep learning" in a discipline, while the less experience teachers will "teach to the test." And though students of the less experienced teachers had better test scores, students who learned from more experienced instructors performed better in subsequent courses. Hoffman cites scientific evidence but the same thing was said, many years earlier, by John Holt, who observed that in a traditional classroom, children learn to play the system, "to manipulate teachers to gain clues about what the teacher really wants. Through the teacher's body language, facial expressions and other clues, they learn what might be the right answer. They mumble, straddle the answer, get the teacher to answer their own question, and take wild guesses while waiting to see what happens."
If we really want to know what students learn, we need to take into account a much wider range of phenomena than how they behave in response to the production of artifacts. Perhaps it's a bit much to measure their pupil dilation, eye gaze, brain activity, blinking, breathing and body temperature, as neuromarketers do. But it shouldn't be too much to expect to be able to map their social and search activity in a learning community, as Google does.
And when we teach, we may not need to take into account everything about the message, the way a political campaign might. We may not, as political consultant Darryl Howard does, "measure everything including the story line, level of the language, images, music." But we should understand, as educators, that learning is much more than mere presentation of facts, that students are learning from everything that goes on around them, and that even if we are not teaching this way, someone -- with perhaps less honorable motives -- is doing it.
SUBSCRIBE TO OLDAILY DONATE TO DOWNES.CA
Web - Today's OLDaily
Web - This Week's OLWeekly
Email - Subscribe
RSS - Individual Posts
RSS - Combined version
JSON - OLDaily
National Research Council Canada
All My Articles
About Stephen Downes
About Stephen's Web
Subscribe to Newsletters
Privacy and Security Policy
Stephen's Web and OLDaily
Half an Hour Blog
Google Plus Page
Huffington Post Blog