Five arguments against explicit teaching

#1 Rote memorisation versus deeper understanding

If you’re a teacher who has begun using a more explicit approach, you’ve probably not spent much time hanging around online following the arguments about explicit teaching. Once you see the benefits in your own classroom, they become obvious, and you’re unlikely to spend much energy worrying about them.

That’s all good. However, you may find yourself working hard, following a path that feels worthwhile and all of a sudden, a school leader or consultant will come along and question what you are doing. At this point, it is worth knowing what they are likely to say. So, let’s take a look.

Argument Number 1: Explicit teaching promotes rote memorisation over deeper understanding.

We all have a sense that there is a qualitative difference between knowing a fact or procedure and some deeper level of understanding. Critics of explicit teaching argue that it is only any good for rote memorising facts and procedures and not for developing this deeper understanding.

Sometimes, they will concede the case for some memorisation and acknowledge that explicit teaching is effective for this. However, even then, they will imply this is a lesser aim and other objectives are more important. I will examine so-called ‘higher order thinking’ in a subsequent post but in this one, I will tackle the idea that something other than explicit teaching is needed to develop understanding.

Empirical evidence

The evidence for this position is weak, at best. In Project Follow Through, students taught with Engelmann’s Direct Instruction—a specific kind of scripted explicit teaching—did just as well or better than their peers in other programs on more complex tasks such as mathematical problem solving and reading comprehension. When David Klahr and Milena Nigam taught students how to control scientific variables via either discovery or direct instruction, not only did fewer learn the principle from the discovery approach, those who did learn from this approach were no better at judging whether science fair posters controlled variables appropriately than those who learned by direct instruction.

Part of the issue is that if you stare too long at ‘understanding’, the concept melts away. When educational psychologists try to measure it, it soon becomes the more concrete ‘conceptual knowledge’ and when Noelle Crooks and Martha Alibali surveyed the literature on this, they found, paradoxically, that demonstrating this knowledge often amounted to trotting out memorised definitions. In contrast, procedural knowledge cannot be as easily regurgitated because it has to be applied to the specific problem a student faces. That’s why I chose to use the ability to apply procedural knowledge to different kinds of problems as my measurement of ‘understanding’ when I completed my PhD experiments.

Despite the common trope among consultants and academics, the evidence suggests that in the case of mathematics, procedural and conceptual knowledge are mutually reinforcing. Does this mean we should only teach procedures and not explain how they work? No, it suggests explaining how they work will help students apply the procedures and applying the procedures will help them understand how they work. Put simply, knowing that 8 x 7 = 56 is not the same thing as knowing what this means, but rather than get in the way of understanding what it means, it complements it.

None of this is at odds with explicit teaching.

Theoretical issues

So, what is understanding? This is a philosophical question and one I have explored before. It links to whether AI can truly ‘understand’ what it is doing and John Searle’s famous Chinese Room thought experiment.

I think there are at least two aspects to understanding. The first is a subjective feeling—we feel we understand something. This could be misleading, of course, but it is unlikely to occur if our working memory is overloaded. We can avoid overloading working memory by keeping the new items students need to process to within the roughly four-item limit of working memory. This can be done either by chunking the content or drawing on schemas students already have in long-term memory and, most of the time, it will be a little of both. Navigating this line is the art of teaching.

Again, this is consistent with an explicit teaching approach.

The second aspect of understanding is more objective: it’s what an observer would recognise as evidence of understanding in someone else, taking us back to the experimental psychologists and their efforts to measure conceptual knowledge. One interesting test in this context is the Turing Test—the test for when a computer has achieved consciousness proposed by Alan Turing. Once a machine reaches the point when we cannot tell its responses from the typed responses of a human, it passes this test. Arguably, AI can now do this and yet I am not convinced it is conscious—but that’s a different matter.

When we examine what this kind of machine understanding looks like, however, it is layers and layers of interconnected knowledge that has been intensively mined from the internet, books, Twitter/X and so on. Whether machines are conscious or not, I see no reason to assume that human understanding does not also consist of knowledge. Distinguishing between knowledge and understanding therefore becomes like distinguishing between houses and bricks.

As I have written before: Knowledge is what we think with. If explicit teaching is the most effective way of passing on knowledge then it is also going to be the most effective way of building understanding.

The influence of constructivism

So, what is the theoretical basis for thinking explicit teaching does not lead to understanding? Partly, it comes from poor experiences at school. We all fail to grasp subjects at times, however they are taught. Learning complex, biologically secondary content is hard. It is easy to blame the teaching style for this and the kind of default explicit teaching that most of us will have experienced—one without frequent formative assessment—can see students drift for days and weeks before a lack of understanding becomes apparent to themselves or the teacher.

No doubt, this fed the prejudice of early progressive educators against explicit teaching and understandably motivated their quest for a better way.

In recent years, that quest has come to be framed in terms of a theory known as ‘constructivism’. This is actually a wide-ranging set of theories that, in some versions, questions the nature of truth and reality. However, the idea that has most currency in education is that our minds ‘construct’ our own representation of the world. Moreover, our minds do not approach reality neutrally and as an empty page, ready to be written on. Instead, we relate new knowledge to what we already know.

In the influential National Academies Press publication, How People Learn by Bransford, Brown and Cocking, this view is explained by reference to a children’s book, Fish is Fish by Leo Lionni. In this book, a frog explains the world above water to a fish but the fish can only interpret it in terms of what it already knows. So the fish imagines people as fish who walk on their tail fins and so on.

In essence, this is useful and true. We do indeed relate new knowledge to what we already know and this is often the source of key misconceptions. However, the inference that we can therefore only understand concepts through direct experience and that we have to be physically active in some way to ‘construct’ this understanding, takes the concept too far and misunderstands human abilities.

If you were living thousands of years ago and a relation told you not to eat a certain plant because it is poisonous, you would understand what that meant. If you read a novel set in 19th century Russia, you can understand what is going on, despite never having visited 19th century Russia. If I explain the solar system metaphor of an atom to you, you can appreciate it, even though you have been stubbornly stuck on earth your whole life and atoms are too small to see.

Yes, direct experience, when feasible, has its advantages, but it’s not essential to understanding a concept.

Humans have evolved an almost miraculous capacity to communicate complex ideas to each other. Novels and metaphors are just two applications of that capacity. Choosing not to use this human superpower, under the misunderstanding that students need to construct everything for themselves, is a waste of everybody’s time.

To come

I will address the following arguments in upcoming posts in this series.

2. Explicit teaching kills student creativity and other higher order skills

3. Explicit teaching kills teacher creativity and autonomy

4. Explicit teaching takes no account of student differences

5. Explicit teaching is demotivating