Lost queens and dodos: some reflections on knowledge and comprehension
In 2016, the passages on the new-look Key Stage 2 ‘Reading’ test caused some controversy, seen by many as being too demanding for too many pupils. They have since been used widely to illustrate the argument for a ‘knowledge-rich’ curriculum in primary schools, and –perhaps more significantly and concerningly – as a rationale for teaching factual knowledge (sometimes called cultural literacy) as a major plank in the teaching of reading.
It has, quite reasonably, been suggested that the vocabulary and references in these texts need not be barriers to understanding if pupils have been taught what they should in the humanities and the sciences. It can be argued that the required background knowledge can and should have been acquired by Year 6, if pupils have been taken through a well-designed curriculum, in which historical, geographical, biological and religious knowledge has been carefully sequenced, layered and revisited, building relevant and rich schemata.
The texts have, therefore, been used to raise a justifiable concern about what was happening in many primary schools, where literacy and numeracy, viewed largely as ‘skills’ only, seemed to take precedence over all other areas of the curriculum. The broader curriculum was being sacrificed to a much narrower focus on test achievement. This is something we would not dispute; shrinking learning about history, geography, religion, myths and science in the primary phase is no good thing.
However, whether one should extend this argument to suggest that it has major implications for our teaching of reading is more open to debate. This post seeks to develop the arguments around this issue.
In his book Why Knowledge Matters, E.D.Hirsch suggests that it is primarily a lack of content knowledge that holds students back from accessing challenging texts and improving their scores in reading tests. Equally, Hirsch blames a lack of vocabulary for students’ inability to improve their reading scores in tests. In some post-Hirsch discourse, this has been turned into a reductive mantra that successful comprehension is essentially only a product of decoding and background knowledge.
It is interesting to note that Hirsch’s arguments are almost entirely based on the idea of improving reading test scores, as the way of judging whether reading has improved. This post seeks to show how that very narrow view of what constitutes reading is a poor way of judging reading development in general. And, in addition, even in its own terms, within the very narrow scope of the reading tests, this view of how to improve ‘scores’ is, in our view, flawed.
Let’s look at the two Key Stage 2 test passages, ‘The Lost Queen’ and ‘The Way of the Dodo’, most frequently used in these debates. Here, supposedly, children need to draw upon a number of ‘big concepts’ from history, around wealth, monarchy, power and lineage, which underlie the telling of the story and which provide its context. The ringed words indicate how these concepts permeate the text and its frames of reference.
According to Hirsch’s way of thinking, accessing this text is largely about being taught these ideas and the associated words first, in a kind of frontloading of essential knowledge. And just as children are said to need lots of historical knowledge to read this text, so in ‘The Way of the Dodo’, it is argued that children need to draw upon their knowledge of zoology and biology.
They also need to understand religious references, including the concept of sacrifice.
They need to make some sense of the historical context.
And they need to have the geographical knowledge to be able to visualise a volcano as something other than just a conical mountain on land.
On the face of it, this looks to be an obvious argument. Clearly, the more general knowledge children have, the more likely they are to be able to navigate the content and the vocabulary of texts. And this is particularly the case if they have not just brushed up against these ideas, but have developed rich schemata, through repeated encounters in a range of contexts.
But there are problems with this argument. However knowledge-rich the curriculum and however well-read the pupil, it simply isn’t possible for every text encountered to offer familiar content and pre-learned vocabulary. And if that were indeed essential for understanding and appreciation, how could texts introduce readers to new worlds and new ideas? How could they teach new knowledge? Knowledge of content is infinite. If the content of a passage is goblins and gnomes, or slavery, or the care of dogs, or glaciers (rather than volcanoes) or Aztec or Greek or native American or Zulu or Indian or Celt or Gaelic civilisation, we can’t teach all this knowledge in advance. Every reading test, on this reckoning, would disadvantage some students who hadn’t been taught that specific content.
An adult literate reader is encountering unknown content and vocabulary all the time; a mark of their literacy is the ability to cope with whatever is thrown at them. In terms of testing reading, where the whole point is to judge how students can cope with unfamiliar texts of a broad nature – how far they have a general competence in reading – it seems counter-logical to teach the content and vocabulary of the text itself in advance, in order to test their reading. That’s a test of knowledge, not reading!
This stance on reading, supported by Hirsch’s views (which, incidentally have been heavily critiqued by educationalists in the US), has also led to a significant and worrying downplaying of other aspects of reading, alongside the over-playing in some practice of the pre-teaching of content, context and vocabulary, whenever pupils encounter a new text.
In a test of reading, the test creators need to bear all of this in mind and choose passages that seem fair in terms of external knowledge required to understand the piece. Arguably, the emphasis in a test should be maximum accessibility in terms of external knowledge and maximum interest in terms of scope for demonstrating understanding and interpretation.
So, one rather obvious question to ask in looking at these two passages chosen for the Key Stage 2 test in 2016 is whether they were well-chosen for a general reading test of the whole cohort of Year 6 pupils. Now they may well be excellent texts for classroom discussion, for reading with a teacher, for group talk and for developing students’ reading strategies but we would argue that – for all kinds of reasons – they are not ideal for a general reading test. The fiction text presupposes knowledge not so much of history, as of English country houses, of the kind that some children might visit with their parents on days out – for instance English Heritage or National Trust properties, with gardens, lakes, follies and monuments. With the second text, one might debate how well written it is and how suitable for a test. While it includes some chatty, colloquial phrases, these seem rather out of keeping with some unnecessarily convoluted sentences and rather old-fashioned phrasing, such as ‘offered themselves up for slaughter’. For a test, is this a great choice?
One other very important angle on this, which in part answers that question, is whether we should extrapolate from what happens in a reading test, to the development of reading more generally. The kind of questioning in these Key Stage 2 tests is mostly of a particular, narrow kind. What has happened to questions like ‘What did you enjoy about…?’, ‘Why do you think the writer…?’, ‘What kind of a story is this…?’, ‘Do you want to read on and why?’, ‘What might this mean?’ and so on. Many of these questions are just as much a part of reading as being able to define words and explain parts of the passage. We would argue that being able to answer these kinds of questions also gives students routes into understanding unfamiliar content and vocabulary, allowing the ‘What do you understand by…’ questions to be answered. (With the ‘Dodo’ passage, one might wonder whether the text itself encourages these kinds of questions or whether it is only eliciting knowledge of difficult words and potentially unfamiliar content).
So tests and testing of reading raise issues that are different from broader consideration of reading development. Tests such as these can be critiqued in their own right.
Let’s move on to consider that broader terrain, imagining that we were thinking about using these texts in the classroom to develop students’ ability to read challenging texts – something that seems to us to be perfectly valid.
With the first passage, instead of starting with unfamiliar content (thrones, monarchs, battles and so on) why not start with the title, ‘The Lost Queen’? It would help you as much, if not more, than knowing what a monument is, to understand what’s at stake in this passage. A student who is used to thinking about the possible meanings of a title will be well served. Who might the lost queen be? A queen in the past? Or the girl whose ancestor lost the battle for the throne? Before reading, during reading and then after reading the passage, coming back to the title with fresh thoughts could lead you into greater understanding.
Another element in this is how much students can infer without knowing everything in advance. Inference is something we all do. It’s a higher order reading skill that allows us to read Henry James and infer characters’ intentions or the author’s use of irony, the subtext of an advertisement or the implications of someone’s account of themselves in a personal statement. We are inferring all the time. In a recent network meeting of secondary English subject leaders, James and the group of teachers explored these ideas by returning to ‘The Lost Queen’ and the ‘The Way of the Dodo’ and conducting a sort of thought experiment, considering how a child without the ‘required’ background knowledge might nonetheless infer or work their way around some of it, by deploying learned strategies – how they might, in fact, make new knowledge for themselves through the act of reading.
For example, a child whose notion of a ‘monument’ was limited to the single prototype of a large, imposing edifice might be confused by the phrase “secret monument”. But if they are used to questioning their own assumptions, and to allowing textual clues to alter their visualisations, then this this could take them quite a long way. (The speech bubble here is not, of course, a realistic representation of what will go through a child’s mind as they read, which will be less conscious and may well be less precise.)
And a child who wasn’t sure about the word ‘ancestors’, and so of its import in discussions of royalty, might infer a sense of its meaning from the context.
A child who hadn’t encountered the word ‘struggle’ in the context of civil war, might nonetheless infer its meaning here.
If they hadn’t come across the word ‘throne’ being used as an abstraction, then they could arrive at this through reasoning.
Or indeed, they could draw on their wider cultural, out of school knowledge. It seems strange that, in all the discussions of this particular text, no-one has suggested that Year 6 children might have heard of Game of Thrones, a series about the battles for power between the rulers of different kingdoms. Might they not be asked to draw on that kind of knowledge to recognise that here a struggle for the throne must be the same kind of thing?
And they could work out an understanding of how ‘symbol’ is used here, within the context.
Again, these speech bubbles are not aiming to be realistic descriptions of what children might say to themselves. However, they actually are very like the ‘think-alouds’ which teachers use when modelling how to make such inferences and how to construct meanings when reading. They represent the sort of explicit processing which teachers draw from children, through skilful questioning in the classroom. And they reflect the sort of implicit, unconscious processing which readers develop through practice.
Meanwhile, in ‘The Way of the Dodo’, a child whose existing schema around volcanoes didn’t encompass seabed eruptions could learn about them there and then, from the text.
This reader is drawing upon other kinds of knowledge – textual, linguistic and literary. They have a deep and implicit understanding of how adjectives, synonyms and metaphors operate, accrued through years of reading and exploring texts, and that gives them the tools to convert their basic geomorphological knowledge into a more advanced kind.
And it is this complex knowledge about how texts work which can get forgotten when reading is reduced to decoding + vocabulary/background knowledge. Scarborough’s famous ‘rope’ is a great visual reminder of this.
In James’s network meeting, this led into a discussion of the other sorts of difficulty which these two texts present to young readers. Focusing on these is just as important as spending time on content knowledge. For example, ‘The Lost Queen’ opens with a syntactically complicated sentence, dotted with prepositions which map the complex spatial arrangements which the reader needs to grasp.
Meanwhile, a challenging variety of tenses and verb forms maps the temporal complexity of the non-linear narrative, which loops around in time and history.
There are also significant complexities in references backwards and forwards (in linguistics called deixis) where you are expected to infer who or what ‘this’ or ‘that’, ‘he’ or ‘she’ is referring to. So, for instance, for inexperienced readers, ‘This was a woman…’ requires some thought about who that ‘this’ is referring to, to work out that it is ‘one of her ancestors’. That is just as tricky as understanding the word ‘monument’. You could understand that word perfectly well and still not be able to work out what’s going on.
Structurally, the reader also has to handle extended passages of description, which divert from the exposition. But the background in paragraph two (moving into explanation of the past) might be more of a problem. Young readers are used to lots of description in storybooks. It might delay the continuation of the story but it’s not hard, in the way that it’s hard to work out why it’s there. ‘Maria led Oliver…’ is both taking us to the point where something important is going to be revealed, and beginning to allow us to piece together the puzzle of what it is.
And here’s where all those other questions not asked in a Key Stage 2 test, come into play. Part of the whole point of this passage is to make us piece it together, isn’t it? It is a little puzzle. The writer doesn’t want us to understand immediately what’s going on and what Oliver is going to find out. She doesn’t inform us of this right at the start, as she could have done. It’s only when the writing is revealed that both Oliver and the reader start to understand why they’re going there and what the implications are for Maria. Oliver is learning along with the reader. These are the very insights we want children to be developing, along the way – why the writer chose to make it a bit complicated, rather than telling us at the beginning that, in the olden days, Maria’s family were kings and queens. We’d want them to recognise that it’s puzzling for the reader, rather than pretending that just by knowing a bit of history, or the words ‘throne’, ‘ancestor’ and ‘monument’ they’d get the point of the passage. The point of the passage is to set up something intriguing and mysterious for the reader.
The language of the passage is richly figurative, packed with metaphor, simile and personification. Again one might want to explore this with children, not only metaphor by metaphor but also in relation to the genre and the effect – the ways in which a sense of uncertainty, mystery and possible danger is suggested, drawing on ways of describing that are typical of mystery stories.
And there is the challenge of being dropped in the middle of a narrative, and of having to accept as given the importance of two unintroduced characters. The reference to ‘the’ (rather than ‘a’) rowing boat adds to this challenge. Are we meant to know about it already? For inexperienced or unconfident readers, these are not trivial obstacles.
Then, the reader isn’t told directly what the two characters are like. Instead, this has to be inferred from a number of adverbial clues.
Meanwhile, ‘The Way of the Dodo’ is made challenging by convoluted, hypotactic sentences, that are just as challenging as the vocabulary and any lack of background knowledge of geography.
The reader needs to be adept at handling idiomatic and figurative expression. ‘Spat out’ is not ‘tier 2’ or ‘tier 3’ vocabulary but in this context it might present difficulties for a reader.
Also, inexperienced readers need to be able to handle the switching between quite formal, ‘tier 2’ language…
…and more informal, colloquial address, resulting in a sort of hybrid register.
This hybrid register might be open to discussion and even criticism. Who is this text intended for, adults or children? How do you know? If for children, is it well judged or not? Does the chatty style of some of it help with the ideas it’s trying to get across? For instance ‘spat out’ is seemingly simple and colloquial but in its context, used metaphorically, it introduces a degree of difficulty in interpretation.
These texts, and the difficulty they presented to Year 6 children, may illustrate to a degree the importance of general and subject-specific knowledge in successful reading, and so the contribution to children’s literacy of a well-designed, broad and ambitious curriculum in the arts, sciences and humanities.
But comprehension is about so much more than this. It is also about constructing meaning from context, and it also draws on the reader’s implicit knowledge and experience of how texts, narratives and language work. The subject English is also, of course, about developing an explicit knowledge and understanding of those things. Importantly, it is also about stepping back and thinking about genre, the choices the writer is making, the impact of the text as a whole and the way we respond as readers, all of which are part of what we might want to assess in reading, and all of which help with that narrower kind of judgement of understanding.
Really, all of this is just a reminder that children don’t become confident, independent ‘comprehenders’ just by being taught lots of general knowledge and vocabulary, even if those things help. And they certainly don’t by answering lots of comprehension questions. Rather, they do so by reading and being read to as much as possible, and by talking about and unpacking what and how they read. They do so through support with thinking about the grammar of texts, their structure, their context, their purposes and their impact, and with responding emotionally and personally, as well as analytically, so that they develop a deep and sophisticated understanding of what it is to make meanings as readers.
To us, the reflections in this post reinforce a few key principles around the teaching of reading.
- Read as broadly and adventurously as possible with pupils.
- Start with response: what do pupils know, think and feel about a text, and how can this lead discussion?
- Keep the big picture in mind at all times: what genre is a text, what is its purpose and who is it written for, what are its effects, what questions does it ask and what are its main ideas?
- Only pre-teach what is helpful to get going with a text: how can the uncovering of meaning, the learning of vocabulary and the piecing together of context be a part of its shared reading?
- Model and practise ‘comprehension strategies’ (such as inference, prediction, summary, activating background knowledge and so on) as an organic part of this shared reading: how can they help to make sense of the text and of the experience of reading it?
- Through conversation about all of these things, explore texts' features – grammatical, structural, literary and so on: how do these parts contribute to the whole?
- Assess all of reading, not just the narrow focus of tests: how can pupils’ broader engagement with and response to texts be recognised and moved on?
- Always read for pleasure: how can all of the above be not about a series of tasks, but about sharing an enjoyable challenge?