Direct Instruction Is Applied Philosophy
I wrote about technical proficiency in the Fall, 2003, issue of Direct Instruction News. I focused on (1) curriculum reform (e.g., how a district might rationally vs. irrationally plan its reading program); and (2) the model-lead-test format as a general communication strategy during initial instruction. In this paper I try to explain how DI curricula are organized and the principles of design that guide that organization.
DI is more than a set of programs. More than a way of organizing instruction. More than one perspective in education. More than mere subject matter. In contrast to almost everything considered "best" and "developmentally appropriate practices" in the field of education (whole language, discovery learning, learning styles, multiple intelligences), DI programs rest on and are guided from the first to the last word by a set of verified design principles. These principles are not fanciful, faddish, unsupported inventions by DI curriculum developers. They are derived from the branch of philosophy called epistemology—theory of knowledge. These principles can be found in the work of Plato (Dialogues), Aristotle (Prior and Posterior Analytics), John Stuart Mill (A System of Logic), and Charles Saunders Peirce ("How to make our ideas clear"). DI is applied epistemology; it uses principles that describe how persons induce (acquire, figure out, construct, "get") general ideas from examples (e.g., the strategy for sounding out a set of words during initial instruction), and then how persons apply this knowledge to new examples (e.g., unfamiliar words). [Bob Dixon wrote elegantly on this issue in Theory of Instruction.]
Logically Technically Proficient Communication
Again, in marked contrast to most of what passes for instruction in education, the aim of Direct Instruction is logically technically proficient communication--so that all students get essential knowledge and get it quickly, and both students and teachers feel continually successful. This section presents what I think are the main principles that guide and are revealed in DI programs—principles that help to account for DI’s reliable effectiveness at teaching almost any student regardless of social class, ethnicity, and family involvement.
Systems of Knowledge
Knowledge systems (such as mathematics, history, physics, literature, how to read) are "out there"—in bodies, books, computers, and other storage and communication devices. These knowledge systems are our species' effort to make sense of and to represent our world. The classical role of teacher is to educate students—from the Latin word educare, to lead forth—out of the cave of ignorance and false belief and into the open air where students, using observation and reasoning strategies (inductive and deductive logic) can come to know how things are. (See Plato, Republic, 29, 514a-521b.)
Knowledge systems consist of and organize 5 forms of knowledge: 1 physical routines and 4 cognitive. These are all and the only the sorts of things we can know and communicate. Everything known boils down to these five forms. Every specific thing and event is an example of these five more general forms. [See Kame’enui and Simmons (1990). Designing instructional strategies; and Engelmann and Carnine (1991). Theory of instruction.]
1. Physical routines include saying sounds, scanning words with your eyes, and writing.
2. Verbal associations are of three kinds:
a. Simple facts. Capital of Massachusetts goes with Boston.
b. Verbal chains. Names of the 13 original states.
c. Verbal discriminations. Students identify documents or statements as representing federalist vs. anti-federalist political positions.
3. Concepts. Examples include red, color, democracy, political system, metaphor, simile, a says ah.
4. Rule relationships, or propositions. For example,
All letters (squiggles) say sounds.
Terrorism never succeeds in its aim.
When rulers violate subordinates' definition of justice, it fosters resistance.
Sounding out is the primary word recognition strategy (not guessing or using context cues).
The more the enemy’s infrastructure is destroyed, the less the enemy resists after defeat.
5. Cognitive strategies. A sequence of steps for analyzing poems, sounding out words, calculating the slope of a line, conducting an experiment, representing a complex process with a theory or diagram/concept map.
Note that each higher form contains all of the lower ones. For example, the strategy for analyzing a poem uses rule relationships ("All Romantic poets believed industrial society is in conflict with nature."); concepts (nature, society, ode, Melancholy), and verbal associations (Keats was a Romantic poet.).
Cognitive Knowledge is Acquired Through Inductive Reasoning
Except for verbal associations (“New fact. Jefferson wrote the first draft of the Declaration of Independence.”), you cannot directly communicate (teach) a concept, rule relationship, or cognitive strategy. You cannot teach students to sound out words in the abstract. You have to use examples. You cannot teach students to analyze poems in the abstract. You have to analyze poems in the here and now. When examples (words to sound out, poems to analyze, math problems to solve) are presented with proper attention to the features in the examples, the sequence in which they are presented, and how the teacher treats them), students induce (figure out, grasp, get) the general concept, rule, or strategy that applies to, incorporates, and is revealed by the examples.
In summary, DI programs aim to move students from superficial or incomplete knowledge of transient and unconnected examples (words, events, and problems that come and go) to knowledge of what is general and enduring—concepts (kinds of things), rule relationships (how kinds of things are connected), cognitive strategies (a. big pictures; e.g., theories and models; and b. problem-solving routines). To do this—to lead students from superficial to general—DI curriculum designers know that one must teach in a way that facilitates students performing certain logical (inductive) operations.
What Logical Operations?
Getting a concept, grasping a relationship, and coming to see the big picture (not just the parts) is a process—a process of reasoning--a sequence of logical steps in which something cognitive is done with the examples. The five logical operations, below, are part of inductive reasoning during initial instruction—how a person induces something general (concept, rule relationship, cognitive strategy) from examples.
Imagine that students read examples and nonexamples of democracies. Or imagine that students are shown examples and nonexamples of granite. By what logical operations do they go from examples to the concept.
1. Examine examples. Look at them. Distinguish features from the whole. DI programs are based on the assumption that students may not know how to do this. Therefore, they must be taught.
Boys and girls. I’ll show you how to examine this rock… First I see that it looks grey. Then I look at it through the magnifying glass. Now I see that the grey is really made of different colored shapes—and each one is a small chip of rock…
2. Note features. Next, the learner focuses on the observed features. In rocks, it might be the shape, color, surface, and hardness of the small chips. In examining societies, it might be language, economic system, time in history, religion, and the degree of control citizens have over their governance.
Again, DI does not assume that students will know how to perform this operation, and so they are taught.
Boys and girls. You see flat slivers of mica, pointed crystals of quartz, and flat-surfaced crystals of feldspar. Touch two examples of each mineral in a sample…. Are there any other kinds of minerals in the four samples I called granite? No. Then these are the three minerals that make granite. Make a numbered list in your notebook and write mica, quartz, and feldspar under the heading Granite….Now look at your samples that I called not granite. Do you see all three minerals--mica, quartz, and feldspar—in each one…?
3. Compare and contrast features across examples. Next, students compare and contrast examples (democracy and granite) and nonexamples (not democracy and not granite). They look from one sample to the other, or better, they compare and contrast the two lists that they made. The teacher might model and then lead students through this step by having students read the lists for each example and nonexample.
Sample 1. Granite. Black mica, yellow feldspar, white quartz.
Sample 1. Not Granite. Yelow feldspar, white quartz.
Sample 2. Granite. Grey mica, white feldspar, pink quartz.
Sample 2. Non granite. Pink quartz, white quartz.
4. Note samenesses and differences. Next students use their comparison and contrast of examples and nonexamples to identify the features that always go with the examples but never go with the nonexamples (mica, feldspar, quartz; high control of governance by citizens). These are the essential samenesses that define the concept. And students identify the features that may or may not go with the examples and nonexamples (language, religion, color, shape). These are the irrelevant features.
All of the societies called democratic have extensive governance by the people. Language, time period, (etc.) differ. So, these can’t be essential. None of the societies called not democracy has extensive governance by the people. This confirms our hypothesis that governance is the defining feature. But language, time period, and economics in nondemocracies are sometimes the same as in democracies. Therefore, not essential.
All of these (a, a, a, a) say ah, even though size and color are different. Shape is the same in each example of ah, but not in any case of non-ah.
5. Draw (construct, discover, state, induce) generalizations that summarize the examination.
The only thing common to all examples called democracy is rule by citizens, and the only thing NOT found in any examples of nondemocracies is rule by citizens. Therefore, rule by citizens MUST be what defines democracy.
This shape (a) says ah. All non-a shapes are not ah. General rules: Shape tells you what sound to say. Recognize the shape.
Of course human beings make inductive inferences from very early on in life (“The thing that smells good delivers the milk.”). But that doesn’t mean they will easily do so when examples have many features, or that they will get the right generalization. Therefore, DI programs explicitly and systematically teach students induction.
Boys and girls. I’ll show you how to induce a general rule. Look at all these triangles. Look at the one on the left. Now, look at the one next to it on the right. How did it change?...It got smaller. Yes, it got smaller. Now look at the next one on the right. How did it change?...It got smaller. Yes, it got smaller. Let’s induce a general rule about how the triangles change from left to right. The triangles get smaller from left to right. Yes, smaller from left to right. Oh, you are so smart.” [Gradually use examples that have more features and that are based on verbal descriptions.]
Inductive reasoning, above, is a set of logical operations for inducing general knowledge (concepts, rule relationships, big pictures, or cognitive strategies) from examples. It is a strategy for inducing something that was unknown (a concept, for example) from something that was known (examples). However, deductive reasoning is a set of logical operations for using or applying (generalizing) concepts, rule relationships, and big pictures, or cognitive strategies, to new examples. It is a strategy for predicting something that is unknown (whether something is an example of a concept) from what is known (the definition of a concept, for example). For instance, once students (during initial instruction) have induced, firmly, that all squiggles shaped like a say ah, they may deduce that new examples with the a shape also say ah. Here are some of the steps, or logical operations in moving from knowledge of the general to predictions about and applications to new specifics.
1. Examine new examples (letters, chunks of rock, descriptions of societies), guided by the general concepts, rules, or theories/schemes (cognitive strategies for making a big picture).
2. Does a new example have the same defining features as are embedded in the general concept, rule, or strategy?...
a. Yes? Then treat the new example as the same as the old ones.
“This says ah.”
“It’s a democracy.”
b. No? Treat the new example differently than the old ones—not as an example of the general case. “Not granite.”
3. Does a new example fit a different general case (monarchy, not democracy; an er verb not an re verb in French; a parentheses with an exponent)? If so, then use knowledge of that other general case to work on it. “That sound is eee.”
Unlike so-called progressive and constructivist pedagogies, DI does not assume that students can make sound deductive inferences without focused instruction. Therefore, DI programs provide that instruction.
Boys and girls. I’ll show you how to make a deductive inference. Remember our rule about how triangles change. What is the rule? Triangles get smaller from left to right. Yes, triangles get smaller from left to right. Well, there is a triangle under this piece of paper and it is on the right. So, what do we know about it? It will be smaller. How do you know? Because it is farther right. Okay, let’s test our deduction….Yes, it is smaller. It fits the rule. What is the rule?...” Then have students do it.
[Gradually use examples whose difficulty matches the ones you will be working on; e.g., making deductions from the theory of revolution in the Declaration of Independence.]
How Do You Communicate in a Logically Technically Proficient Fashion So That Students Easily Perform the Inductive and Deductive Operations?
The guiding ideas or design principles and the instructional methods in DI programs are derived from and are perfectly consistent with its theory of knowledge--the set of inductive and deductive logical operations by which persons acquire and apply knowledge. Following are some of the main design principles and instructional features.
1. Assume that students are not likely to figure things out (make inductive and deductive generalizations on their own). Therefore, you must pay attention to every bit of knowledge they must learn, how clearly you communicate, and what they are getting.
2. Teach virtually every verbal association, concept, proposition, and cognitive strategy directly (not round about), systematically (step by step with attention to details) and explicitly ("The definition of democracy is...").
3. Each form of cognitive knowledge (verbal association, concept, rule relationship or proposition, and cognitive strategy) can be taught effectively with its own communication format. Use the same format when teaching all examples of the same knowledge form. For example…
New concept. Democracy. Democracy is a political system in which citizens rule. Say that definition… This is an example of democracy…. And this is an example of democracy… This is NOT an example of democracy. This IS an example of democracy… Now look at this example… Is this democracy?...How do you know.”
4. Pre-teach whatever students need in order to get what you are trying to teach.
a. Teach letter-sound correspondence before sounding out words (because the latter involves the former).
b. Teach counting before addition and subtraction.
c. Teach definitions of terms before analyzing documents that use those terms.
5. Teach the subskills used in logical operations. And teach these as part of a cognitive strategy called “reasoning.” That is, teach students how to examine examples and nonexamples, how to note features, how to write what they note, how to compare and contrast and discover samenesses and differences, how to draw inferences (concept, rule, big picture—theory, model), how to apply the general case (concept, rule, big picture) to new examples.
6. Introduce a big idea—concept, rule, model/diagram—to organize thinking from example to example and from day to day.
Remember the rule. When governments no longer secure what citizens consider their unalienable rights, this delegitimizes the government, and this fosters opposition movements.
7. State what you are going to teach and what students will be able to do--the
Do-objectives Statements such as “Students will appreciate rhyme,” or “demonstrate rhyme,” or “make rhyme” are not do-objectives. A do-objective would be:
Teacher models how to rhyme with at. r/at, m/at, s/at. Teacher says, “Your turn to rhyme with at. rrrr…, mmm…, sss….” Students correctly rhyme (r/at, m/at, s/at) within 3 seconds.
The do-objective tells you EXACTLY what to teach--to rhyme at with rrrr, mmm, and sss in response to the teacher saying, “Your turn to rhyme with at.”
The do-objective also tells you exactly what to test to determine acquisition (rhyme at with rrr, mmm, sss); fluency (students rhyme quickly); and generalization (students also rhyme at with fff, hhh, and vvv).
8. Teach in a logically progressive sequence.
a. Teach general examples before unique ones. Letter-sound correspondence for a, m, s, f, r, e before v, ing, x.
b. Teach elements before compounds, parts before wholes.
Vocabulary before reading texts that contain the words.
Counting forward before addition.
c. Teach what is useful now before what is useful useful later.
9. Teach so that inductive inferences are easy to make. Specifically,
a. Use examples that reveal/contain the features relevant to a concept, rule, or strategy.
b. Use a range of examples that have features in the population of examples OUTSIDE of instruction (to avoid stipulation errors).
Examples of Blue
c. Juxtapose examples to make sameness and difference obvious.
Not glerm ---
Not glerm 0
10. Direct students’ attention to relevant events.
“Put your finger under the…" "Look at the…" Watch me…"
11. Use prompts to highlight important features. Change the circle on d to an oval to distinguish it from the similar shape of b.
13. Test immediately and, after students have worked on many examples, test again (delayed acquisition test).
14. Strategically integrate into a larger whole what is learned earlier in a lesson or in a series of lessons. For example, teach students to use knowledge of rhyme schemes, meter, figures of speech, and symbolism in the strategy for analyzing poems.
In the earliest examples we have (e.g., Plato’s portrayals of Socratic dialogue, Aristotle’s analyses of arguments), technical precision in communication is seen as necessary to move learners from shallow and erroneous belief to enduring knowledge. This precision is attained by organizing communication so that it fosters the logical operations (reasoning) learners must perform in order to get (induce) and apply (deduce) knowledge. In this sense, DI is one of the few forms of instruction remaining that not only maintains classical aims (mastery of traditional systems of knowledge) but as well provides classical means (logically proficient communication).