WORDS 14 GRADES
WORDS 14 GRADES
Now that I am retired, I realise even more how problematic ranked grading of students' work is, even though I have been acquainted with the extensive research on this since the 1960s, which demonstrated that even in maths it was impossible to eradicate subjectivity, and that the detailing of assessment criteria for such ranking is anything but scientific despite the apparent specificity of quantitative metrics. Blind marking of the same paper by the same marker could result in an A grade one year and a Fail the next year.
The first instance of this type of numerical grading that I have come across is that done by a minor painter and art 'theorist' of the 17th Century, when he applied grades to the eminent painters of the Renaissance. Roger de Piles, in The Principles of Painting with a Balance of Painters, 1708 creates a table with four assessment criteria, and gives each painter marks out of 20 for each criterion. He titles this table:
"A CATALOGUE of the names of the most noted Painters and their Degrees of Perfection, in the Four principal Parts of Painting, supposing absolute Perfection to be divided into twenty Degrees or Parts"
He then divides the Table into four columns:
COMPOSITION;
COLOUR;
DRAWING;
EXPRESSION
In his text he writes:
"The method I have taken is this: I divide my weight into 20 parts or degrees. The 20th degree is the highest, and implies sovereign perfection; which no man has fully arrived at. The 19th is the highest degree that we know of which no person has yet gained. And the 18th is for those who, in my opinion, have come nearest to perfection."
(EG Holt, ed. 1958, A Documentary History of Art, Volume II, pages 184 and 185)
You may be surprised to learn that a now relatively unknown painter called Le Brun scored 56 out of 80, beating Michelangelo who scored a measly 37 out of 80, and Rembrandt who scored just 50, and Leonardo at 49/! I don't mean to say that our assessments of these painters are necessarily more objective than those of Roger de Piles. Both assessments are subjective, and require discursive arguments that acknowledge that subjectivity to be viable. But de Piles attempts to make his unargued assessment of each criterion sound objective and absolute by using numbers.
The first, according to Wikipedia, written examinations were introduced at Cambridge in 1747. As to be expected, wikipedia is wrong:
"Trinity College Cambridge may have held the first written examinations in Europe, ...if so it must be dated to 1560 rather than 1702 - with the proviso that the themes once written, were probably read out before the Examination Fellow." (C. Stray 'From oral to written examination: Oxford, Cambridge and Dublin 1700-1914'. in History of Universities 2005, 20 (2): 76-130
This ushered in the possibility of cheating, which led to one of my witty colleagues quipping that we teachers invented cheating - you can't cheat during an oral exam.
Grading of candidates dates back at least to the 18th Century at Cambridge:
"It was at about the time of Bentley's arrival, in the 1700s, that the beginnings can be detected of what became known as the Senate House Examination, and later the Mathematical Tripos. This was a University degree examination whose history through the eighteenth century is one of increasingly fine differentiation of grading. Since the sixteenth century, the highest-achieving BAs of each year had been listed in an order of merit (the Ordo Senioritatis); the others were listed separately in college groups." I n 1710-11 the higher men were listed in two groups. First Tripos and Second Tripos, and though the terminology changed, this division into two classes persisted in subsequent years. From 1747-8 the list was printed. From 1753 the first class was divided into two, and this was the origin of the distinction between wranglers and senior optimes; the second class consisting of junior optimes. Together these classes represented the three classes of honours, the other candidates being known as 'hoi poloi' (the masses)." (C. Stray ibid.)
In 1797 William Farrish at Cambridge University introduced quantitative grading of papers in his subjects. (K. Hoskins 'The Examination, Disciplinary Power and Rational Schooling' in History Of Education Volume Eight Number Two 1979 pages 135 to 146, Quoted in Neil Postman Technopoly, 1993, Page 13).
I am not denying that judging actions and substances is a universal and necessary activity. This is food and that is poison. This guy knows how to fix a car and that guy is a crap mechanic. What I am saying here Is that grading at this quantitative detail, pretending to pseudo-precision with numbers, is part and parcel of the ideology of institutionalising the hierarchical worth of humans. Whether it is merit that places one on a higher rung of the ladder, or competitive edge, or just one's birth, matters not in practice. But this type of 'grading' is meant to prove through the apparent precision and objectivity of numbers that assigning a number to a person establishes the 'objective' worth of that person
Another type of numerical grading is meant to prove that it is genetics that establishes value and worth: the I.Q. Test. Neil Postman says S.J.Gould makes three points about this:
[the first is reification, the converting of an abstract idea Into a thing but] ”…There is no such ‘ thing’ as intelligence. It is a word not a thing and a word of a very high order of abstraction.. But if we believe it to be a thing like the liver then we will believe scientific procedures can locate it and measure it."
Second problem is ranking which requires a criterion for assigning individuals to their place in a single series. If we use objective-sounding numbers to rank individuals' natural intelligence in a single series we therefore assume intelligence is not only thing but a single thing located in the brain and accessible to the assignment number. It is as if beauty inhered in the size of a woman's breasts. Then all we would have to do is measure breasts and rank each woman accordingly and we would have an objective measure of beauty.
The third problem is that we will thus have defined our question in a restricted way. But this would go unnoticed because of scientism's mystique of numbers, which are the ultimate test of objectivity. By using numbers here we give even ourselves the illusion of objectivity when in fact the test is subjectively framed.
A fourth problem not mentioned by Postman here is that the IQ 'number' is not a quantity, but signifies a ratio: it expresses the difference of an individual score compared to the 'average' score on these tests for someone of the same age. Thus, if one was 10 years old but scored the same as an average 15 year old, you would then divide 15 by ten, giving you 1.5, then multiply by 100, to get 150. We cannot, however, assume that an aveage 15 year old is 50% more intelligent than an average 10 year old, but this number makes it appear so. This tells us nothing about an actual quantitative measure, even if this was possible. My guess is that Einstein, Niels Bohr, et al. had a sepcific intelligence that was mch higher than their ‘IQ’ scores would indicate.
A fifth problem is that it tests mainly the speed with which certain types of abstract numerical and word problems can be solved. Given enough time, many if not most could finish the test reasonably successfully. This test does not measure the ability to cope with very complex symbolic problems such as those in higher maths, symbolic logic or quantum physics. There may be a correlation between a high IQ score and ability in these areas, but it is still not known whether this is a necessary condition. It certainly does not seem to be a sufficient one, as most high IQ people I have met have not shown much beyond being very quick off the draw, often too quick, and thus making more mistakes than a 'slower' thinker.
And as for wisdom, there does not seem to be any relationship between quick off the draw and wisdom.