Last changed 14 Aug 2013 ............... Length about 5,000 words (33,000 bytes).
(Document started on 2 April 2012.) This is a WWW document maintained by Steve Draper, installed at You may copy it. How to refer to it.

Web site logical path: [] [~steve] [best] [this page] [triad 1] [triad 2] [triad 3]

Reading, discussing, writing

By Steve Draper,   Department of Psychology,   University of Glasgow.

"Reading maketh a full man, conference a ready man, and writing an exact man."

This aphorism and the essay it comes from suggest that studying, in fact I would say Higher Education (HE), rests on three activities: reading, discussing, and writing. Since Bacon is often cited as the first to publish an explicit scientific method, we should consider whether this formula is a general educational one and not limited to essay-based Humanities subjects. It comes from an essay by Francis Bacon in 1625, and in 1753 Samuel Johnson wrote another essay elaborating on it (local copies). Together they make the case for the importance of each, and how omitting any one leads to weaknesses: all three are required for rounded learning. We might say that they correspond to receiving ideas, interacting about / with ideas, and generating one's own detailed idea.

Currently it may constitute a relevant and insightful critique of HE where there is far too little discussion by students of ideas. The measure of this is the number of minutes per day a given student is actually speaking about some intellectual idea. (Listening to discussion may have some other value, but does not count at all under this heading, as Johnson's essay makes clear if you look at it with this question in mind.)

If we were to take this as a serious educational rule, then for each course we need to consider an even division in times spent on each of reading, discussing, and writing; and also an equal weight of assessment for each.

(A rather different explanation, more psychological than educational, would explain discussing vs. writing in terms of extraversion/intraversion. Susan Cain on the power of introverts (20 mins). "There's zero correlation between being the best talker and having the best ideas." Correct: but the point from Johnson is that everyone is better for having both skills, not either/or; and regardless of the personal disposition measured by the personality dimension.)

Finally, you might like to consider Bayard's (2007) arguments about the importance and inevitability of discussing (i.e. talking about) books you haven't read.

Paying more attention to discussion

Bacon and Johnson were both stressing the equal importance of the three, but I suspect HE today needs to focus particularly on discussion. My argument is, that when people had to travel and thus live with each other in order to study, discussion didn't have to be particularly arranged for. Today universities are over-crowded and many students commute to them. Peer discussion is increasingly unlikely to happen naturally.

Perhaps a modern equivalent of discussing might be video games; or online question banks. Like discussion, they are interactive (not pure self-generation) and reactive, making a complementary learning activity to reading and writing. This argument then amounts to the ideas that discussing is powerful for learning because:

Another reason to focus on discussion is the (neo-)Vygotskian idea that dialogue is especially important in learning: precedes each conceptual development. A particular form of this view is explored by Sfard (2008).

There is however rather direct evidence on the importance of discussion from Treisman's work (1983, 1992; Fullilove & Triesman, 1990). He studied the differences in habits of students from different minorities at U.C. Berkeley; found the Asian Americans spent a lot of time discussing work together and the others did not; introduced interventions to get the others to discuss, and had a big success in subsequent grade improvements. One might however try to summarise his success in different theoretical ways e.g. (Tinto-like) social and academic integration; discussion (this page); a combination of collaborative learning and problem-based learning (what Chinn & Martin (2005) call it).

Thinking = writing

Confucius discusses the role of "thinking": shouldn't thinking (which we could call "reflection" if it makes us feel better) be part of education?

"I used to sit alone thinking about this and that. Sometimes I even forgot my meals or bedtime. Still I gained very little. Later I shifted to reading omnivorously, but I did not benefit a great deal either. At long last I came to see that reading in a mechanical way without using my brains was no use. On the other hand, if thinking is divorced from the reality and no due attention is paid to reading, one will continue to feel puzzled by many things. One should constantly review what he has learned and combine reading with thinking. In thus making use of the theories one has learned to guide his thought and help analyze the problems at hand, progress will be achieved." (Confucius)

I suggest that writing, from the standpoint of learning, plays the part of thinking: the effort of writing forces careful thinking. Johnson says nothing about anyone else having to read what you write: the point is the precision and order that writing demands of you, and plenty of time to do it (unlike in conversation). In that case, Confucius (two thousand years earlier) was also pointing out the complementary requirements for reading and thinking = writing in successful learning.

Calculations as the counter-part of writing

In essay based subjects, writing is how we find out what we think; or perhaps, work out more implications and consequences for our initial view. In calculation-based subjects (STEM subjects like maths and physics) perhaps calculations are the corresponding activity. (By "calculation" I include arithmetic, algebra, and mathematical proofs.)

You certainly see physicists who are arguing or thinking scribble down rough calculations both for themselves, and as part of a discussion with others. Calculations are pieces of reasoning, usually beyond what we can hold in our heads with accuracy and confidence: and that is just why writing prose is so valuable to thinkers in other disciplines.

In STEM subjects, students are typically required to do regular "problems" i.e. calculations, just as in essay-based subjects students are generally required to write regularly. But they need all three activities, as Feynman said in 1963 (in the Preface to his published physics lectures):
"I think, however, that there isn't any solution to this problem of education other than to realize that the best teaching can be done only when there is [....] a situation in which the student discusses the ideas, thinks about the things, and talks about the things. It's impossible to learn very much by simply sitting in a lecture, or even by simply doing problems that are assigned.
But in our modern times we have so many students to teach that we have to try to find some substitute for the ideal. ..."

Manual skills, procedural learning

So far, this page is about book learning (of declarative / conceptual ideas), in contrast to the learning of practical skills (as in triad 1 and triad 2). For that, reading, discussing, writing seem to be a fundamental trio. When reading the author's intention is dominant, in discussing both parties' ideas are important, and writing is where the learner's own goals and ideas drive the activity. Similarly, the trio correspond to receiving, negotiating, generating ideas; and to interaction with an expert, peers, and oneself (reflection). The directions of transfer are inwards (to the learner), in both directions, then outwards. We might equally refer to it as listening, reacting, then thinking and doing. (And when Confucius speaks of teaching, it seems to have been dialogic and not essentially similar to reading or lecturing.)

However if we consider learning a skill, e.g. cooking or surgery, the trio is transformed into: watching it being done, trying it out under supervision, and finally teaching someone else to do it. In this case, the middle phase is interacting primarily with the materials rather than with a peer, and with the teacher who has to respond to what the pupil does (rescuing their errors) rather than following a planned course of their own. The final phase of teaching a skill is typically where the student for the first time takes control of the overall activity and also adds to their manual actions reflection and an attempt to articulate the activity: corresponding to writing for a desk-based learner. Hence the surgeons' slogan "See one, do one, teach one". Or observing, attempting, directing.

Confucius and other connections

In "The uncommon reader", Alan Bennett portrays various consequences of becoming a serious reader. Discussion doesn't become a large feature, possibly because his protagonist is royal. However the structure of his piece is that what is begun by starting to read inevitably ends by becoming a writer, so he too seems to be following the structure articulated by Bacon and Johnson.

Confucius doesn't mention writing at all in this way, but talks about thinking which seems to play the same role. Confucius talks (at least in one translation of the Analects) of thinking, reading, and teaching. ("Keeping silent and thinking; studying without satiety, teaching others without weariness: these things come natural to me." [Analect 7.2])

If we can equate thinking and writing with respect to their benefit to learning, then this is writing, reading, and teaching; or doing it oneself (by thinking or writing), seeing it (by reading another's expression of it), and teaching (i.e. discussing it with another person). This suggests a close analogue between Bacon's aphorism for conceptual learning, and the practice-oriented surgeon's aphorism for skill learning: "see one, do one, teach one".

Another angle on read / discuss / write may be the framework offered in Fonseca & Chi (2011). They compared the observed learning gains for learning:

Summary points

Beyond read/discuss/write

In Bacon's and Johnson's times, it was still reasonable to think that there were a limited number of important books (the classics) and that to become educated was to read all of them personally. All of the above argument is about such personal learning based on reading. Bayard (2007) however shows us that, although we haven't admitted this to each other yet, it is absolutely impossible for any of us to read everything important in our lifetimes. Even if it once was, the nature of the world of learning in which we now live is otherwise. An almost inevitable consequence is that we must equip ourselves to talk about books we haven't read.

In practice, as opposed to in what we care to be aware of and announce, this has always been a core academic skill. It comes out classically in making an argument and citing authorities in support. When you do that, you are not giving the proof but trusting (and requiring your own readers to trust) the cited authority. Such abbreviation is essentially like the traditional exercise of making a précis in that an original piece of writing is replaced by a summary plus an explicit pointer of where to go for more detail. And this too is a core intellectual skill: deciding what summary of a large argument or report is adequate for intellectual purposes. It is quite different from the trio discussed on this page, and needs to be taught and practised differently.

However it also, like the trio above, does correspond to a deep feature of human knowledge and learning: the socially distributed nature of our shared knowledge. As Putnam (1975) showed, even for material facts which can be grounded in independently observable facts, we almost always only know part of what is to be known but we know where to go if and when more details are required. This is essentially like knowing the précis, not the whole thing; like being able to talk about Shakespeare, but not having seen or read most of his plays.

It overlaps with the topic of "discussion"; but leads to a different deep feature of studying.

Chi's Passive, Active, Constructive, Interactive

Chi (2009), Fonseca & Chi (2011) propose a framework of learner involvement that goes: Passive, Active, Constructive, Interactive.

But isn't that essentially Reading, Writing, Discussion? (at least if we skip "active" as the kind of notes you make while reading).

If this mapping is correct, then her view (and data) implies that peer discussion (her Interactive mode) has bigger learning effects than the rest.

(On the other hand, perhaps discussion is what is most missing in ordinary education,and so has the biggest experimental effect; but that directly comparative experiments would show that all are equally important.)

Extended to social digital media?

This is Steve Wheeler's thought: to apply a version of Bloom's taxonomy to student behaviour as modified by social digital media, and end up with something like the read-discuss-write spectrum.

This is his two page blog piece on the idea
as prompted by his earlier two page blog piece on
student lurking and loafing.

His diagram

His stages have the same developmental direction from reading through discussion to writing. And the pyramid narrows: quite rightly: for everything you read, you only discuss some, and write about still fewer. But he has 5 stages, not my 3.

Reinterpreting Wheeler's pyramid

I think his stages can be thought of as generated by crossing read/discuss/write with the learner having a passive/reactive stance, vs. an active (self-directed) one; thus connecting to Wheeler's lurking/loafing theme. I.e. if active, the learner wants to learn about the topic and acts to achieve this; but if passive, the learner is responding to a friend raising the topic, a teacher putting something in front of them. Labelling this "lurking or loafing" is perjorative: actually a teacher-centered stance that presupposes that there is a clear joint task (dictated by the teacher) and any failure to join in whole-heartedly is the learner's fault, not the fault of the person imposing the activity. In some situations that is true; and a teacher might argue it is true of school. But it is NOT true of learning in general, where although this is little acknowledged, we all spend a lot of time lurking and waiting for something interesting. That is what newspapers and magazines are specifically designed for; RSS feeds; browsing journals (as opposed to looking up a specific cited article). Lurking is in fact the mark of someone passively open to inputs from elsewhere. A fully active, self-directed person would walk down the street, ignore the fact that everyone was staring up at the same thing, and walk right under the collapsing building because it wasn't their self-chosen goal to research that at that moment. In other words, we get a lot of our information, some of it the most important possible, not by being "active" but by lurking.

This would give the following stages. I show them as bullet points, so development / time now goes down the page not up:

  1. Reading, reactive: watching, lurking, browsing
  2. Reading, mixed: follow links from something you read through
  3. Reading, self-directed: Tracking down a specific piece; Reading it all.
  4. Discussing, reactive; expressive: sharing, retweeting, liking, favouriting. Mentioning in passing something you read.
  5. Discussing, reactive: Reply to email, add a comment to a blog i.e. some new content generated but only because prompted by someone/thing else.
  6. Discussing, self-directed: Commenting at some length. Feeling the need to discuss something at length, finding a willing or at least acquiescent partner for this.
  7. Writing, self-directed; expressive: writing for oneself e.g. diary, notes, to work out what you think.
  8. Writing, self-directed: Creating, repurposing.
  9. Writing, reactive, instrumental: writing for a specific audience; "curating".

Reactive Mixed Self-directed
Read a b c
Discuss d e f
Write i h g

An interesting feature is that the sequence begins passively and becomes more active in the reading phase; the discussion phase is on the whole balanced w.r.t. active/passive. The best discussions have two motivated people; and we are often infuenced by whether others bring up a topic, which can amplify our initial interest. With writing, this is inverted: first writing is typically for oneself (expressive), while the final writing stage is doing it only for others' sake (instrumental): researching an audience, expressing it differently to suit them (or, to spin it downwards, selling your soul to write formulaic novels or advertising copy). And in between these is writing to first create and then convince a new audience.

See Peter Elbow's (1987) memorably titled paper on this point: "Closing my eyes as I speak: An argument for ignoring the audience" which begins with the quotation "Very often people don't listen to you when you speak to them. It's only when you talk to yourself that they prick up their ears." And the obverse of someone who ignores their audience in order to deliver what they in fact value the most is a reader who is too limited in what they want to engage with the best (most transformative) material: "If you travel with us you will have to learn things you do not want to learn in ways you do not want to learn".
[Doris Lessing, from a letter replying to a reader who had been seriously disturbed by reading one of her novels. Quoted in Alan Yentob's "Imagine" TV programme on Doris Lessing, broadcast Tues 27 May 2008, 10:35pm on BBC1]

Basically, the more self-directed the writer is, the more passive the reader must be; and the more self-directed the reader is, the more passively reactive the writer must be.

In fact it can be argued that, important as discussion is as part of rounded learning, reading and writing are both fundamentally NOT dialogic: not improved but only diminished by interaction. Proust for instance said that the best way to get to know someone was not to meet them but to immerse yourself in their writings; and Proust was in many ways a highly sociable person. Thus, while this page is written from a suspicion that there is too little discussion in HE today for optimal learning, an opposite mistake may be admiring social media and thinking that solitary reading and writing are less essential. Bacon's whole point was that you need all three, and that a deficiency in any one leads to characteristic pathologies, which he sketches.

I think this whole framework would equally apply, not only to writing, but to drawing, painting, designing coffee mugs, creating pottery, ....

Summary notes, as a table

Relationships amongst versions of read/discuss/write, and to virtues
Read Discuss Write
Bacon Reading Conference Writing
Confucius Reading Teaching Thinking
Chi (2009) Passive (active) Interactive Constructive
Physics etc. Reading (equations as well as prose) Discussing Calculating
The inner activity Knowing (understanding) Developing answers to critical challenges Self-correction of facile success in argument
(neo-Vygotsky?) Interact with: An expert Peers Oneself
Whose goal?: The author's Both The learner's
Directionality: Inwards 2-way Outwards
Directionality: Receive Dialogue Transmit
Directionality: Listening Reacting Directing
Surgical analogue(4) Teacher models Teacher coaches Teacher scaffolds
Surgical analogue(3) Watching Attempting to imitate it (under supervision) Taking charge (directing)
Surgical analogue(2) Watch one Try one Teach one
Surgical analogue See one Do one Teach one
Virtue (if done the right amount) A full man A ready man An exact man
If done too little, then to compensate you need: Cunning to conceal your ignorance Quick wittedness A big memory
If done too much Sloth Affectation "The humour of the scholar"
Benefit(2) Delight Ornament Ability
Benefit(2.2) Amuse yourself "Talk well" for others Taking decisions, disposition of business.
Benefit(3) ? Grace Method


Web site logical path: [] [~steve] [best] [this page]
[Top of this page]