One of my most memorable learning experiences that has served me well for decades, and that I actually recall most days of my life, occurred during a teacher training session early in my teaching career. We had been set the task of giving a two-minute lecture on something central to our discipline. Most of us did what we could with a slide or two and a narrative to match in a predictably pedestrian way. I remember none of them, not even my own, apart from one. One teacher (his name was Philippe) who taught sports nutrition, just drew a picture of a banana. My memory is hazy on whether he also used an actual banana as a prop: I'd like to think he did. For the next two minutes, he then repeated 'have a banana' many times, interspersed with some useful facts about its nutritional value and the contexts in which we might do so. I forget most of those useful facts, though I do recall that it has a lot of good nutrients and is easy to digest. My main takeaway was that, if we are in a hurry in the morning, not to skip breakfast but to eat a banana, because it will keep us going well enough to function for some time, and is superior to coffee as a means of making you alert. His delivery was wonderful: he was enthusiastic, he smiled, we laughed, and he repeated the motif 'have a banana!' in many different and entertaining ways, with many interesting and varied emphases. I have had (at least) a banana for breakfast most days of my life since then and, almost every time I reach for one, I rememember Philippe's presentation. How's that for teaching effectiveness?

But what has this got to do with educational technologies? Well, just about everything.

As far as I know, up until now, no one has ever written an article about bananas as educational technologies. This is probably because, apart from instances like the one above where bananas are the topic, or a part of the topic being taught, bananas are not particularly useful educational technologies. You could, at a stretch, use one to point at something on a whiteboard, as a prop to encourage creative thinking, or as an anchor for a discussion. You could ask students to write a poem on it, or calculate its volume, or design a bag for it. There may in fact be hundreds of distinct ways to use bananas as an educational technology if you really set your mind to it. Try it - it's fun! Notice what you are doing when you do this, though. The banana does provide some phenomena that you can make use of, so there are some affordances and constraints on what you can do, but what makes it an educational technology is what you add to it yourself. Notwithstanding its many possible uses in education, on balance, I think we can all agree that the banana is not a significant educational technology.

Parts and pieces

Here are some other things that are more obviously technological in themselves, but that are not normally seen as educational technologies either:

• screws
• nails
• nuts and bolts
• glue

Like bananas, there are probably many ways to use them in your teaching but, unless they are either the subject of the teaching or necessary components of a skill that is being learned (e.g. some crafts, engineering, arts, etc) I think we can all agree that none of these is a significant educational technology in itself. However, there is one important difference. Unlike bananas, these technologies can and do play very significant roles in almost all education, whether online or in-person. Without them and their ilk, all of our educational systems would, quite literally, fall apart. However, to call them educational technologies would make little sense because we are putting the boundaries around the wrong parts of the assembly. It is not the nuts and bolts but what we do with them, and all the other things with which they are assembled, that matters most. This is exactly like the case of the banana.

Bigger pieces

This is interesting because there are other things that some people do consider to be sufficiently important educational technologies that they get large amounts of funding to perform large-scale educational research on them, about which exactly the same things could be said: computers, say. There is really a lot of research about computers in classrooms. And yet metastudies tend to conclude that, on average, computers have little effect on learning. This is not surprising. It is for exactly the same reason that nuts and glue, on average, have little effect on learning. The researchers are choosing the wrong boundaries for their investigations.

The purpose of a computer is to compute. Very few people find this of much value as an end in itself, and I think it would be less useful than a banana to most teachers. In fact, with the exception of some heavily math-oriented and/or computer science subjects, it is of virtually no interest to anyone.

The ends to which the computing they perform are put are another matter altogether. But those are no more the effect of the computer than the computer is the effect of the nuts and bolts that hold it together. Sure, these (or something like them) are necessary components, but they are not causes of whatever it is we do with them. What makes computers useful as educational technologies is, exactly like the case of the banana, what we add to them.

It is not the computer itself, but other things with which it is assembled such as interface hardware, software and (above all) other surrounding processes - notably the pedagogical methods - that can (but on average won't) turn it into an educational technology. There are potentially infinite numbers of these, or there would be if we had infinite time and energy to enact them. Computers have the edge on bananas and, for that matter, nuts and bolts because they can and usually must embody processes, structures, and behaviours. They allow us to create and use far more diverse and far more complex phenomena than nuts, bolts, and bananas. Some - in fact, many - of those processes and structures may be pedagogically interesting in themselves. That's what makes them interesting, but it does not make them educational technologies. What can make them educational technologies are the things we add, not the machines in themselves.

This is generalizable to all technologies used for educational purposes. There are hierarchies of importance, of course. Desks, classrooms, chairs, whiteboards and (yes) computers are more interesting than screws, nails, nuts, bolts, and glue because they orchestrate more phenomena to more specific uses: they create different constraints and affordances, some of which can significantly affect the ways that learning happens. A lecture theatre, say, tends to encourage the use of lectures. It is orchestrating quite a few phenomena that have a distinct pedagogical purpose, making it a quite significant participant in the learning and teaching process. But it and all these things, in turn, are utterly useless as educational technologies until they are assembled with a great many other technologies, such as (very non exhaustively and rather arbitrarily):

• pedagogical methods,
• language,
• drawing,
• timetables,
• curricula,
• terms,
• classes,
• courses,
• classroom rules,
• pencils and paper,
• software,
• textbooks,
• whiteboard markers,
• and so on.

None of these parts have much educational value on their own. Even something as unequivocally identifiable as an educational technology as a pedagogical method is useless without all the rest, and changes to any of the parts may have substantial impacts on the whole. Furthermore, without the participation of learners who are applying their own pedagogical methods, it would be utterly useless, even in assembly with everything else. Every educational event - even those we apparently perform alone - involves the coparticipation of countless others, whether directly or not.

The point of all this is that, if you are an educational researcher or a teacher investigating your own teaching, it makes no sense at all to consider any generic technology in isolation from all the rest of the assembly. You can and usually should consider specific instances of most if not all those technologies when designing and performing an educational intervention, but they are interesting only insofar as they contribute, in relationship to one another, to the whole.

And this is not the end of it. Just as you must assemble many pieces in order to create an educational technology, what you have assembled must in turn be assembled by learners - along with plenty of other things like what they know already, other inputs from the environment, from one another, the effects of things they do, their own pedagogical methods, and so on - in order to achieve the goals they seek. Your own teaching is as much a component of that assembly as any other. You, the learners, the makers of tools, inventors of methods, and a cast of thousands are coparticipants in a gestalt process of education.

This is one of the main reasons that reductive approaches to educational research that attempt to isolate the effects of a single technology - be it a method of teaching, a device, a piece of software, an assessment technique, or whatever - with the intent of generalizing some statement about it cannot ever work. The only times they have any value at all are when all the technologies in question are so hard, inflexible, and replicable, and the uses to which they are put are so completely fixed, well defined, and measurable that you are, in effect, considering a single specific technology in a single specific context. But, if you can specify the processes and purposes with that level of exactitude then you are simply checking that a particular machine works as it is designed to work. That's interesting if you want to use that precise machine in an almost identical context, or you want to develop the machine itself further. But it is not generalizable, and you should never claim that it is. It is just part of a particular story. If you want to tell a story then other methods, from narrative descriptions to rich case studies to grounded theory, are usually much more useful.