AQE, Belfast Newsletter, Ben Lowry, Carla Lockhart MLA, Department of Education, DUP, GL Assessment, Parental Alliance for Choice in Education, Peter Robinson., Peter Weir MLA, Professor Peter Tymms, Stephen Elliott
Aristotle, Daniel Willingham, Direct Instruction, Dr Hugh Morrison, J D Hirsch, ludwig Wittgenstein, Max Bennet, Neuroscience & Philosophy: Brain Mind & Language, Peter Hacker, Stephen Elliott, Steven Pinker, TES, University of Virginia
Why E.D. Hirsch’s particular brand of “science” is powerless to challenge direct instruction
Dr Hugh Morrison (The Queen’s University of Belfast [retired])
The cover page of the Times Education Supplement (TES) quotes the distinguished American educationalist E D Hirsch’s claim that “there is no scientific basis for direct instruction.” Given the high regard in which Hirsch is held by educational traditionalists, there will be widespread dismay that one of their own is invoking science to attack a traditional pedagogical technique that can see off any progressivist model when it comes to raising the educational standards of poor children. (Readers who google the words “direct instruction” will see why this classroom approach is so important to traditionalists.) Hirsch clearly appreciates that the reasoning set out in his new book may not be received with universal acclaim: “To offend everybody is one of the few prerogatives left to old age.” The good news for proponents of direct instruction everywhere is that the “science” Hirsch appeals to makes no sense.
The basis of Hirsch’s TES “scientific” attack is the field of cognitive science. To convince his readers that his book represents “consensus science” he has invited two prominent cognitive scientists, Steven Pinker and Daniel Willingham, to “blurb” the book. Now cognitive scientists hold that psychological attributes like thought, understanding, memory, meaning, and so on, are internal processes associated with the human brain/mind. According to cognitive scientists the mind/brain is a self-contained realm where computations are performed on mental “representations.” However, one of the towering figures of 20th century thought, Ludwig Wittgenstein, regarded this type of thinking as deeply misconceived. He wrote that: “The confusion and barrenness of psychology is not to be explained by calling it a “young science”: its state is not comparable with that of physics, for instance, in its beginnings. … For in psychology there are experimental methods and conceptual confusion.” In their 2003 book Philosophical Foundations of Neuroscience Max Bennet and Peter Hacker use Wittgensteinian reasoning to attack cognitive neuroscience’s central claims. In the remainder of this essay the “consensus science” which informs Hirsch’s claims is undermined using the writings of Bennett and Hacker. The reader needs neither a background in cognitive psychology nor a grounding in philosophy to appreciate immediately the validity of Wittgenstein’s “conceptual confusion” claim; a healthy dose of common sense will reveal immediately the error at the heart of cognitive science.
While it is clear that thinking would be impossible without a properly functioning brain, the claim that brains can think or that thinking takes place in the brain ought to be supported with scientific evidence. No such evidence exists. To mistakenly attribute properties to the brain which are, in fact, properties of the human being is to fall prey to what Bennett and Hacker refer to as the “mereological fallacy.” (Mereology is concerned with part/whole relations and the fallacy goes all the way back to Aristotle.)
“Psychological predicates are predicates that apply essentially to the whole living animal, not to its parts. It is not the eye (let alone the brain) that sees, but we see with our eyes (and we do not see with our brains, although without a brain functioning normally in respect of the visual system, we would not see). So, too, it is not the ear that hears, but the animal whose ear it is. The organs of an animal are part of the animal, and psychological predicates are ascribable to the whole animal, not its constituent parts” (pp. 72-73).
Cognitive scientists often refer to brains “thinking,” “knowing,” “believing,” “deciding,” “seeing an image of a cube,” “reasoning,” “learning” and so on.
“We know what it is for human beings to experience things, to see things, to know or believe things, to make decisions … But do we know what it is for a brain to see … for a brain to have experiences, to know or believe something? Do we have any conception of what it would be like for a brain to make a decision? … These are all attributes of human beings. Is it a new discovery that brains also engage in such human activities?” (p. 70)
In the words of Wittgenstein (1953, §281): “Only of a human being and what resembles (behaves like) a living human being can one say: it has sensations; it sees, is blind; hears, is deaf; is conscious or unconscious”. If the human brain can learn, “This would be astonishing, and we should want to hear more. We should want to know what the evidence for this remarkable discovery was” (Bennett & Hacker, 2003, p. 71). It is important to appreciate the depth of the error committed here. When the claim that the brain can think is called into question, this doesn’t render valid the assertion that brains, in fact, cannot think.
“It is our contention that this application of psychological predicates to the brain makes no sense. It is not that as a matter of fact brains do not think, hypothesise and decide, see and hear, ask and answer questions; rather, it makes no sense to ascribe such predicates or their negations to the brain. The brain neither sees, nor is it blind – just as sticks and stones are not awake, but they are not asleep either” (p. 72).
One gets the clear impression from the cognitive science literature that understanding or remembering are inner processes. Wittgenstein, while accepting that without a properly functioning brain one couldn’t learn, nevertheless teaches that understanding is something attributed to the whole person, and not the brain. When a teacher asks a pupil what she thinks, the pupil expresses her thoughts in language. Were it not for the pupil’s language skills, the teacher couldn’t ascribe thoughts to her. Since brains aren’t language-using creatures, how can it make sense to ascribe thoughts to a brain?
While cognitive scientists may protest that the brain’s ability to make connections while it (the brain) is learning, is visible from the PET or fMRI images of the brain, scientific writing should always show restraint:
“But this does not show that the brain is thinking, reflecting or ruminating; it shows that such-and-such parts of a person’s cortex are active when the person is thinking, reflecting or ruminating. (What one sees on the scan is not the brain thinking – there is no such thing as a brain thinking – nor the person thinking – one can see that whenever one looks at someone sunk in thought, but not by looking at a PET scan – but the computer-generated image of the excitement of cells in his brain that occurs when he is thinking.)” (pp. 83-84).
In Neuroscience & Philosophy: Brain, Mind and Language (2007, p. 143), Bennett and Hacker write:
“But if one wants to see thinking going on, one should look at the Le Penseur (or the surgeon operating or the chess player playing or the debater debating), not his brain. All his brain can show is which parts of the brain are metabolizing more oxygen than others when the patient in the scanner is thinking.”
In order to see off Hirsch’s ill-founded claims, advocates of direct instruction can appeal to no less a thinker than Aristotle. Around 350BC he wrote:
“to say that the soul (psyche) is angry is as if one were to say that the soul weaves or builds. For it is surely better not to say that the soul pities, learns or thinks, but that a man does these with his soul.”
The solution to the single test issue for grammar schools is as clear as it is simple: how defensible is the practice of using two tests to determine grammar school entry?
In Northern Ireland, grammar schools usually admit pupils using one of two tests: the vast majority of “state” grammar schools use the AQE test to select pupils, and the vast majority of Catholic grammar schools use the GL test. A small number of schools engage in a process which has come to be called “dualling,” whereby pupils who have taken either the AQE test or the GL test (or both) are considered eligible for entry. Moreover, the algorithm for assigning children to a single score scale using scores from these very different tests (see below) is shrouded in mystery. Despite its importance to parents applying to the school, I can find no such algorithm on the website of any “dualling” school. However the National Foundation for Educational Research (NFER) does caution against misuse stating “Care should be taken when considering the change in percentile rank of one test-taker on different occasions or on different tests” See http://www.nfer.ac.uk/research/centre-for-assessment/standardised-scores-and-percentile-ranks.cfm
In what follows, I argue, on behalf of test-takers that “dualling” is unlikely to be defensible. Given the recent entry of the Presbyterian Church in Ballymena into this debate, it will be useful to frame the case that “dualling” is unlikely to be defensible using the Church’s concern that testing only adds to the “stress already inherent in moving from primary to post-primary school.” My central preoccupation will be the radically differing approaches to test anxiety taken by the AQE and GL tests.
The Transfer Test, abolished by Direct Rule ministers and Sinn Fein when that Party took up the education portfolio, consisted of two tests, each of one hour duration, taken on separate days and assessing mathematics, English and science. Children, who might, for example, have had a sleepless night before taking Transfer Test 1, had the opportunity to improve their score on the second test. In these circumstances, a sleepless night might only impact adversely on one of the two assessments. Moreover, by holding the tests on different days, children, through access to their teachers in the period between tests, had opportunities to address shortcomings in their knowledge and skills which might have come to light in Transfer Test 1, in time for Transfer Test 2. I will now argue that while the AQE test, by design, seeks to reduce levels of test anxiety below those associated with the Transfer Test, the GL test represents a backward step in addressing test anxiety when compared with the Transfer Test.
GL’s guidance literature reveals that children are involved in almost two hours of testing (with a comfort break in the middle) at one sitting. (It should be noted no GCSE examination – taken by much older pupils – exceeds 90 minutes.) In the GL framework there is no opportunity to address shortcomings between test occasions because the assessment in its entirety takes place on one day. A sleepless night has the potential to impact adversely on all aspects of the child’s test performance. Added to this, the children have to carefully transfer each of their answers to a multiple-choice grid because the GL test is marked by “optical mark recognition” software.
While the GL test has greater potential for stress and anxiety than its predecessor, the AQE test was designed to improve on the Transfer Test in this respect. The AQE test involves three one hour tests and seeks to improve on the Transfer Test’s two-test structure in respect of test anxiety. The first and second tests are separated by a fortnight, and test three is taken one week after test two. There is plenty of opportunity to address misconceptions with one’s teacher as one goes from test to test. The impact of a sleepless night is greatly reduced because the child’s best scores on two of the three tests are used to compute his or her published score. The process discards the child’s lowest score. Finally, children simply write their answers (and rough work) on the test paper, with the need to transfer each answer to a grid obviated entirely.
Why should it matter that one test (GL) has decreased the number of test occasions previously available under the Transfer Test, while the other (AQE) has moved to increase this number? Why has AQE gone to such lengths to reduce the potential for test anxiety to levels below those associated with the Transfer Test? If it is accepted that the AQE and GL tests differ significantly in respect of their approach to test anxiety, what implication could this have for the practice of “dualling”?
The answer is that differences in respect of test anxiety have significant measurement implications. Even for older test-takers, there is a well-established correlation (on a scale from -1 to +1) of approximately -0.4 between test score and test anxiety (see Wildemuth (1977): Test anxiety: An extensive bibliography (ERIC/TM Rep. 64). Princeton, NJ: Educational Testing Service).
Questions about the defensibility of “dualling” arise because the two tests used to assign children to a single rank order, measure differently as a consequence of their differing approach to test anxiety. The Ballymena Presbytery is right to emphasise the centrality of minimising pupil stress. This has been at the heart of the AQE project since its inception. Any effective “dualling” algorithm must take account of the differing approaches of the AQE and GL assessments to test anxiety. Because no such algorithm exists, “dualling” can play no role in attempts to address the concerns raised by the Presbytery. The solution is as clear as it is simple: parents must demand that all grammar schools in Northern Ireland adopt the same selection test.
Chairman, The Parental Alliance for Choice in Education
The above article was published by The Ballymena Guardian in response to a concern expressed by the Ballymena Presbytery on Thursday October 2nd, 2014 Read it here http://www.ballymenaguardian.co.uk/articles/news/42455/presbyterys-grave-concern-over-school-tests/
Questions will arise from parents considering the reduction of anxiety concern for their children when contemplating using “dualling”