Wednesday, September 23, 2009

Facts versus skills

> How do you define facts versus skills?

Tracy, you get to the core of a deep issue very quickly. Let me try a brief response, with the admission that a longer response may be necessary.

My response is this: there isn't really a distinction between facts and skills; what we call 'facts' and what we call 'skills' are the very same thing.

However, what people mean when they say 'facts' is something very different, and it is this sense of 'facts' to which I am responding in this essay.

In this sense of 'facts', a fact is something that it is propositional, it is declarative knowledge, it is expressible as a sentence, it is stored atomically (as a 'sentence in the brain') in memory, it is privileged (that is, it is a proposition that is known to be 'true').

Now, we can cause people to remember these sorts of things, and to recite them back when needed, and storage of these sorts of things can be useful in some cases, because they can serve as a guide to action.

A good example of this is a pilot's pre-touchdown checklist. This is a set of statements which need basically to be recited and acted upon ("landing lights - check - flaps - check - throttle back - check - wheels down - check..."). Explicit declarative memory can serve as a guide to action, and is useful in high-risk situations (like landing an aircraft) where you can't make a mistake, or in novice situations (like installing a software application for the first time, or just beginning to use mathematics).

But this is the exception and not the rule. Acting as though all knowledge - or any significant proportion of it - resembles declarative knowledge is a fundamental misrepresentation. Treating learning as the storage and accumulation of facts is, in general, a mistake.

It's confusing, because the same bit of knowledge can be represented as both a skill and a fact. Take driving a car, and 'knowing which pedal is the accelerator'.

Pretty clearly, we can say "we need to know which pedal is the accelerator" in order to drive a car. But what does that mean?

If we represent it as a declarative memory, then the process of driving a car is described as "remembering which pedal is the accelerator, selecting it from the field of vision, remembering how to press on the accelerator,... etc." But if we represent it as a skill, then the process of driving the car is "accelerating..."

Now, in the latter case, we did not dispense with the "knowledge of which pedal is the accelerator" - but at the same time, we did not appeal to it explicitly, nor did we need to. The knowledge is implicit and contained within the much more complex (and non-declarative) skill of 'accelerating'.

Why is this distinction important? Well, when we are teaching, when we are learning, and when we are evaluating, we need to be clear about which form of knowledge we are working with: declarative memory, or skill. Because, if it's declarative memory, then nowing which pedal is the accelerator will not be sufficient for knowing how to accelerate - you need to know that you have to press it, how hard to press it, when to ease off, how to steer while you are doing it, not to run over the family dog, etc...

If we tried to represent 'accelerating' as declarative knowledge, we would have to specify each of these steps. This would work only in the most basic instances of accelerating. Such a description might be useful to get the skill of 'accelerating' off the ground. But, note:

- the declarative does not actually constitute learning how to accelerate - knowing the sentence isn't the same as knowing how to accelerate, and the learning of 'how to accelerate' does not consist of learning the sentence at all, but rather, of the learning, through practice, the actual skill and habit of how to accelerate

- an evaluation of whether a person knows how to accelerate is in no way supported through the succesful answering (or non-answering) of questions related to the declarative knowledge

- teaching that is focused on the recollection of 'facts' (as declarative statements) may improve a person's ability to respond to questions about those statements, but does not in any significant way support learning that fact (as a constituent of the larger and more complex skill)

This is pretty easy to demonstrate with the example of driving, or landing an airplane, because these are obviously skills, and cannot be reduced to declarative knowledge.

But what I am saying is that knowing is a skill, just like driving, and that there are constituent skills to knowing - skills like literacy, learning, prioitizing, evaluation, planning and acting.

The learning of these skills - and the application of these skills - does not involve the learning or use of declarative memory. When (for example) reading, you do not go through some step-by-step process of decoding, you navigate the text in the same sort of way you navigate a road when you drive. When you evaluate a piece of writing for fallacies, you do not run through a list of fallacies in your mind, you recognize a fallacy through he same ineffable process that you would use to recognize a dog running across the road.

Declarative knowledge - the list of how to spell words, the addition and multiplication table, the rules of grammar, the list of fallacies - might help you get started, but they're just that: a starting point. Not the best, or only, or uniquely meaningful starting point. Just a leg-up into the more complex process. A teacher's aid. Scaffolding.

Hope that helps.

6 comments:

  1. Thanks Stephen for explaining your definition. This of course raises some further questions:

    How do you know that the opponents of "21st century skills" are using the same definition of facts as you give here? For example, I just looked up the word "fact" on google and got the following link: http://www.answers.com/fact
    The first definition this gives of fact is "Knowledge or information based on real occurrences: ..." The second is "Something demonstrated to exist or known to have existed".

    I can't see anything in this definition about facts needing to be declarative knowledge, or expressible as a sentence, or stored atomically. For example, while I can't tell you in declarative knowledge terms how far I need to depress the brake pedal in my car to get the car to stop in a certain distance on a dry asphalt road, I know this in the sense that this is knowledge based on real occurrences, and it is something that has been demonstrated to exist (subject of course always to the assumption that my memories have not been meddled with by aliens, and I'm not unknowingly actually living in a virtual simulation and etc).

    Also, just out of curiousity, how do you know that a fact is storied atomically in the brain? (I don't know much about how memory is stored).

    ReplyDelete
  2. Another thing, sorry I just noticed it. You say:
    "But what I am saying is that knowing is a skill, just like driving, and that there are constituent skills to knowing - skills like literacy, learning, prioitizing, evaluation, planning and acting.

    The learning of these skills - and the application of these skills - does not involve the learning or use of declarative memory."

    But in this essay you are declaring many facts. For example, you said earlier "A good example of this is a pilot's pre-touchdown checklist. " And after the bit I quoted you immediately go on to say "When (for example) reading, you do not go through some step-by-step process of decoding, you navigate the text in the same sort of way you navigate a road when you drive."

    If we don't need to use declarative memory to knowledge, or to use such skills as evaluation, why are you using so much declarative knowledge in your essays?

    ReplyDelete
  3. Conscious and sub-conscious use of knowledge seems to be an important part of this discussion. I suggest reading the article and some of the chapter the can be found at these urls, as they should help.

    http://socialbrain.rsablogs.org.uk/tag/candle-problem/
    http://books.nap.edu/openbook.php?record_id=9853&page=31

    Communication is the indirect attempt to transfer concepts between the non-directly connected mental systems of us humans. Without some form of declarative knowledge, communication fails. However, those that might push the non-declarative angle of teaching are referring to setting a person up for exploration based learning. By applying skills they gain declarative knowledge.

    Declarative knowledge can be used and practiced with so it goes from conscious use to sub-conscious use, a skill. When this happens it is common for the declarative knowledge to be forgotten in its declarative form. Also, some facts are never put into declarative form, or only one or two declarative forms. Depressing a peddle while driving, for instance, has more than one way to gauge when it is sufficiently depressed. If you can describe a success condition for depressing a particular peddle, then you have a declarative form of "how far I need to depress the brake pedal". From there conditional statements can be added about the distance and conditions to modify the declarative form similar to a math, or logical, equation. Since most people do this via "skills" the equation is not always simplified into a consciously recognizable declarative form of knowledge.

    I've also found that it is generally an iterative process going back and forth between skill and fact forms where-in a person learns the most.

    ReplyDelete
  4. > How do you know that the opponents of "21st century skills" are using the same definition of facts as you give here?

    I can't be sure, of course, since there's always a difference in meaning when two different people use the same concept. I look for their usage in sentences and their descriptions of attributes.

    If their use of the word 'fact' is _not_ as I describe, so much the better for me - but then they are challenged on some of the implications of this, as (for example) a transport theory of communication or transaction theory of learning require facts to be delarative sentences (propositions, concepts, etc).

    > For example, I just looked up the word "fact" on google and got the following link:... The first definition this gives of fact is "Knowledge or information based on real occurrences: ..." The second is "Something demonstrated to exist or known to have existed".

    The problem with dictionary definitions is that they are vague and somethings circular. To say that a fact is 'knowledge' helps us only if 'knowledge' is not defined as 'facts'. The invocation of 'real' or 'exiting' are actually appeals to concepts less well understood than 'fact'. My definition is considerably more precise than what you'll see in the dictionary because the dictionary definitions are inadequate.

    > While I can't tell you in declarative knowledge terms how far I need to depress the brake pedal in my car to get the car to stop in a certain distance on a dry asphalt road, I know this in the sense that this is knowledge based on real occurrences, and it is something that has been demonstrated to exist

    What sense of the word "exist" applies to "how far I need to depress the brake pedal"? It can't be 'exist' in the sense of 'physical object' because a distance isn't a physical object. Does it 'exist' because it's 'know;edge'? That just creates circular reasoning.

    This is the problem when you get into these sorts of discussions using words and language - as soon as we try to be clear about what we mean, we bounce from one thing to the next, never getting closer to our goal...

    > Also, just out of curiousity, how do you know that a fact is storied atomically in the brain? (I don't know much about how memory is stored).

    *If* a fact in the brain is a sentence, it is stored atomically because a sentence is atomic - that is, it has a distinct and concrete instantiation.

    It is (partially) because I don't think that facts are atomic that I don't think they can be sentences, and therefore, not sentences in the brain.

    > in this essay you are declaring many facts.... If we don't need to use declarative memory to knowledge, or to use such skills as evaluation, why are you using so much declarative knowledge in your essays?

    I have to use declarative knowledge in my essay, because my essay is expressed in language, and declarative knowledge is the only thing that can be expressed in language.

    It follows that I think that my essay is, at best, a stepping stone that will help you (and others) to come to understand the concepts of knowledge and learning for yourselves, but that it is *not* the knowledge I am trying to impart.

    I'm trying to show you what knowledge and learning look like, to point you to it, to maybe even demonstrate it - but I can't just state it as a bunch of facts that you'll read and understand and, as a result, know. Knowledge of my essay is not the same as knowledge of knowledge and learning.

    ReplyDelete
  5. If their use of the word 'fact' is _not_ as I describe, so much the better for me - but then they are challenged on some of the implications of this, as (for example) a transport theory of communication or transaction theory of learning require facts to be delarative sentences (propositions, concepts, etc).

    What's a transport theory of communication, or a transaction theory of learning?

    And why must a fact be a declarative sentence even within a transport theory of communication or a transaction theory of learning?

    The problem with dictionary definitions is that they are vague and somethings circular.

    Indeed. This is an epistomological problem, how do we agree on what a word in a language means? I don't know, I can't remember how I did it myself when I was a baby.
    Ignoring the epistomological problem for the moment, the value is that dictionaries allow some ideas about how other people might be thinking about words. Your definition may be far more precise than what you see in a dictionary, but if other people are not using the word "facts" in the precise sense that you are, your definition is misleadingly precise. Precision is not the only value in language.

    What sense of the word "exist" applies to "how far I need to depress the brake pedal"?

    It exists in the sense that from my experience if I depress the brake pedal a certain amount the car stops in a certain distance on a road with a certain surface concept, subject of course to the assumptions that my memories are reasonably accurate reflections of reality. This is not a precise definition of exist, I agree. It is always possible that the next time I depress the brake pedal on my car at a certain speed on a certain road surface the car will fail to stop in a certain distance, say because the breaks have failed. But just because I don't have a precise definition of "exist" doesn't mean that the word is useless or meaningless.

    I know that philosophers spend pages and pages of ink, or more recently electrons, agnoising over how we know what we know and what can we know, and what do we mean by words like existence and so forth. I think these discussions are interesting and worthwhile, but unless you advocate abandoning any attempt at an education system until the matter is finally resolved I think there's a point where we should stop worrying about it and get on with the more immediate question of what we should learn and what the best ways of teaching it are.

    And on a more practical level, how many philosophers finish writing an article on the epistomological problem, hop into their car to head home, and then sit there in the drivers' seat agonising because they can't be sure how they know that the brakes are likely to work when they push the brake pedal? (Substitute other agonising for non-driving philosophers, such as failing to make themselves a cup of coffee because they can't be sure how they know that applying heat to the kettle will cause it to boil).

    This is the problem when you get into these sorts of discussions using words and language - as soon as we try to be clear about what we mean, we bounce from one thing to the next, never getting closer to our goal...

    At least however by these discussions we can raise our awareness of the bits we are unclear about.

    ReplyDelete
  6. As for your response to my question as to why you are using declarative language, your response illustrates the importance of teaching facts (in a looser definition that includes things that may not be declarative knowledge) as part of teaching. You say that you use declarative knowledge because declarative knowledge is the only thing that can be expressed in language, and you are trying to lead me, and others, to understand the concepts of knowledge and learning.
    And that I think is why the core knowledge people argue that curriculae should explicitly include the teaching of content, including facts. Not because facts, in the broader definition, are the only things we want to teach, but that these facts are how we access the deeper knowledge that we want imparted, we can't separate content from knowledge, and of course when we want to teach knowledge about true things (as opposed to fictional things like the plot of King Lear) we need to teach facts in the broader sense of the word facts.

    ReplyDelete

I welcome your comments - I'm really sorry about the moderation, but Google's filters are basically ineffective.