Calling a spade a spade: Mathematics in the new pattern of division of labour
CCALLING A SPADE A SPADE:MATHEMATICS IN THE NEW PATTERN OF DIVISION OFLABOUR
ALEXANDRE V. BOROVIK
The man who could call a spade a spadeshould be compelled to use one.It is the only thing he is fit for.
Oscar Wilde Introduction
The growing disconnection of the majority of the population from mathematicsis increasingly difficult to ignore.This paper focuses on the socio-economic roots of this cultural and social phe-nomenon which are not usually mentioned in public debates. I concentrate onmathematics education, as an important and well documented area of interactionof mathematics with the rest of human culture.New patterns of division of labour have dramatically changed the nature and roleof mathematical skills needed for the labour force and correspondingly changedthe place of mathematics in popular culture and in mainstream education. Theforces that drive these changes come from the tension between the ever deepeningspecialisation of labour and ever increasing length of specialised learning requiredfor jobs at the increasingly sharp cutting edge of technology.Unfortunately these deeper socio-economic origins of the current systemic crisisof mathematics education are not clearly spelt out, neither in cultural studies nor,even more worryingly, in the education policy discourse; at the best, they are onlyeuphemistically hinted at.This paper is an attempt to describe the socio-economic landscape of mathemat-ics education without resorting to euphemisms. This task imposes on the authorcertain restrictions: he cannot take sides in the debate and therefore has to refrainfrom giving any practical recommendations. Also it makes necessary a very cleardisclaimer:
The author writes in his personal capacity. The views expressed do notnecessarily represent the position of his employer or any other person,organisation, or institution.
The last pre-publication version, 11 December 2014. c (cid:13) a r X i v : . [ m a t h . HO ] D ec ALEXANDRE V. BOROVIK The new division of labour
It’s the economy, stupid.
James Carville A word of wisdom from Adam Smith.
Discussion of mathematics edu-cation takes place in a socioeconomic landscape which has never before existed inthe history of humanity.This, largely unacknowledged, change, can be best explained by invoking AdamSmith’s famous words displayed on the British £
20 banknote, Figure 1:
Figure 1
The words on the banknote:
The division of labour in pin manufacturing (and the great increase inthe quantity of work that results) are, of course, a quote from Adam Smith’s
The Wealth of Nations . They are foundon the very first page of Chapter I of Book I with the now famous title
Of TheDivision of Labour : One man draws out the wire; another straights it; a third cuts it; a fourthpoints it; a fifth grinds it at the top for receiving the head; to make thehead requires two or three distinct operations; to put it on is a peculiarbusiness; to whiten the pins is another; it is even a trade by itself to putthem into the paper; and the important business of making a pin is, inthis manner, divided into about eighteen distinct operations.
And Adam Smith comes to the conclusion: . . . they certainly could not each of them have made twenty, perhaps notone pin in a day; that is, certainly, not the two hundred and fortieth,perhaps not the four thousand eight hundredth part of what they are atpresent capable of performing, . . . It’s the economy, stupid.
According to Wikipedia, this phrase, frequently attributed to BillClinton, was made popular by James Carville, the strategist of Clinton’s successful 1992 presiden-tial campaign against George H. W. Bush.
ALLING A SPADE A SPADE 3
By the start of the 21st century, the ever deepening division of labour hasreached a unique point in the history of humankind when 99% of people havenot even the vaguest idea about the workings of 99% of technology in their im-mediate surrounding—and this applies even more strongly to technological uses ofmathematics, which are mostly invisible.Every time you listen to an iPod or download a compressed graphic file fromthe Internet, extremely sophisticated mathematical algorithms come into play. Asmartphone user never notices this because these algorithms are encoded deep insidethe executable files of smartphone apps. Nowadays mathematics (including manytraditional areas of abstract pure mathematics, such as number theory, abstractalgebra, combinatorics, and spectral analysis, to name a few) is used in our everydaylife thousands, maybe millions, of times more intensively than 50 or even 10 yearsago. Mathematical results and concepts involved in practical applications are muchdeeper and more abstract and difficult than ever before. One of the paradoxes ofmodern times is that this makes mathematics invisible because it is carefully hiddenbehind a user friendly smartphone interface.There are more mobile phones in the world now than toothbrushes. But themathematics built into a mobile phone or an MP3 player is beyond the under-standing of most graduates from mathematics departments of British universities.However, practical necessity forces us to teach a rudimentary MP3/MP4 technol-ogy, in cookbook form, to electronic engineering students; its mathematical contentis diluted or even completely erased.2.2.
A few more examples.
New patterns of division of labour manifest them-selves at every level of the economy.2.2.1.
A consumer.
25 years ago in the West, the benchmark of arithmetic compe-tence at the consumer level was the ability to balance a chequebook. Nowadays,bank customers can instantly get full information about the state of their accountsfrom an app on a mobile phone.2.2.2.
A worker in the service sector.
How much arithmetic should a worker ata supermarket checkout know? And they are being replaced by fully automatedself-checkout machines.2.2.3.
A worker in an old industry.
Even in the pre-computer era, say, in the 19thand the first half of 20th centuries consumers were increasingly ignorant of the fullextent of technological sophistication used in the production of everyday goods.In relation to mathematics that meant that buyers of ready-to-wear clothing, forexample, were likely to be unaware of craft-specific shortcuts and tricks of geometryand arithmetic used by a master cutter when he made a template for a piece ofclothing. In the clothing industry nowadays, cutters are replaced by laser cuttingmachines. But a shirt remains essentially the same shirt as two centuries ago; givenmodern materials, a cutter and a seamstress of yesteryear would still be able toproduce a shirt meeting modern standards (and millions of seamstresses are stilltoiling in the sweatshops of the Third World). What a 19th or 20th century cutterwould definitely not be able to do is to develop mathematical algorithms which, afterbeing converted into computer code, control a laser cutting machine. Design andoptimisation of these algorithms require a much higher level of mathematical skillsand are mostly beyond the grasp of the majority of our mathematics graduates.
ALEXANDRE V. BOROVIK
A worker in a new industry.
Do you need any mathematical skills at all forsnapping mobile phones together on an assembly line? But production of microchipsis highly automated and involves a very small number of much better trained andeducated workers. Research and development in the area of microelectronics (andphotonics) is of course an even more extreme case of concentration of expertise andskills.2.2.5.
International division of labour.
It is easy to imagine a country where not asingle person has a working knowledge of semiconductor technology and productionof microchips. What for? Microchips are something sitting deep inside electronicgoods imported from China—and who cares what is inside? Modern electronicgoods usually have sealed shells, they are not supposed to be opened. Similarly,one can easily imagine a fully functioning country where no-one has mastered, say,long division or factorisation of polynomials.2.3.
Social division of labour.
In the emerging division of intellectual labour,mathematics is a 21st century equivalent of sharpening a pin.The only difference is that a pin-sharpener of Adam Smith’s times could betrained on the job in a day. Development of a mathematically competent workerfor high tech industries requires at least 15 years of education from ages 5 to 20.
It is this tension between the ever-increasing degree ofspecialisation and the ever-increasing length of specialisededucation that lies at the heart of the matter.
At this point we need to take a closer look at social division of labour. Braverman[8] emphasises the distinction between the social division of labour between differentoccupational strata of society and the detailed division of labour between individualworkers in the workplace.
The division of labor in society is characteristic of all known societies;the division of labor in the workshop is the special product of capitalistsociety. The social division of labor divides society among occupations,each adequate to a branch of production; the detailed division of labordestroys occupations considered in this sense, and renders the workerinadequate to carry through any complete production process. In cap-italism, the social division of labor is enforced chaotically and anarchi-cally by the market, while the workshop division of labor is imposed byplanning and control. [8, pp. 50–51]
It is the new workplace, or “detailed”, division of labour that makes mathemat-ics redundant in increasingly wide areas of professional occupation. Meanwhile thelength-of-education constraints in reproduction of a mathematically skilled work-force lead to mathematics being single out not only in workplace division of labour,but also in social division. And, exploiting the above quote from Braverman, it isthe “chaotic and anarchic” nature of social division that leads to political infightingaround mathematics education and paralyses education policy making.The rest of my paper expands on these theses. One point that I do not mentionis the division of labour within mathematics; this is an exciting topic, but it requiresa much more specialised discussion.
ALLING A SPADE A SPADE 5 Politics and economics
The issue of new patterns of division of labour has begun to emerge in politicaldiscourse. I give here some examples.The book by Frank Levy and Richard Murnane
The New Division of Labor [21],published in 2004 and based on material from the USA, focuses on economic issuesviewed from a business-centred viewpoint. Here is a characteristic quote:
In economic terms, improved education is required to restore the labormarket to balance. [. . . ] the falling wages of lower skilled jobs reflect thefact that demand was not keeping up with supply. If our predictions areright, this trend will continue as blue-collar and clerical jobs continue todisappear.Better education is an imperfect tool for this problem. The job marketis changing fast and improving education is a slow and difficult process.[21, p. 155].
Elizabeth Truss, a Conservative Member of Parliament and Secretary for theEnvironment (who until recently was an Undersecretary of State in the Departmentfor Education), not long ago published a report [24] where she addressed the issueof the “hourglass economy” in the context of education policy.
The evidence suggests increased polarisation between high skilled andunskilled jobs, with skilled trades and clerical roles diminishing. Longstanding industries are becoming automated, while newly emerging in-dustries demand high skills. Formal and general qualifications are themain route into these jobs. At the top level MBAs and international ex-perience is the new benchmark. Despite popular perception, the middleis gradually disappearing to create an ‘hourglass economy’. [24, p. 1]
In the next section, we shall return to the “hourglass economy” and the “hour-glass” shape of the demand for mathematics education to different levels of students’attainment. Meanwhile, I refer the reader to the views of numerous economists con-cerning “job polarisation” (Autor [5], Goos et al. [17]), “shrinking middle” (Abeland Deitz [1]), “intermediate occupations” and “hourglass economy” (Anderson[4]). The same sentiments about the “disappearing middle” are repeated in morerecent books under catchy titles such as Tyler Cowen’s
The Average is Over [12];they are becoming part of the
Zeitgeist . Although their book is optimistic, Brynjolf-sson and McAfee [9] emphasise the way in which the application of the know-howin the upper half of the hourglass causes the hollowing out of the “neck”.It is instructive to compare opinions on job polarisation and its impact on edu-cation coming from opposite ends of the political spectrum.Judging by his recent book [18, Chapter 14], Alan Greenspan focuses on the toppart of the hourglass: [W]e may not have the capability to educate and train students up to thelevel increasingly required by technology to staff our ever more complexcapital stock. The median attainment of our students just prior to WorldWar II was a high school diploma. That level of education at the time,with its emphasis on practical shop skills, matched the qualifications,by 1950s standards, for a reasonably skilled job in a steel mill or auto-assembly plant. [. . . ] These were the middle income jobs of that era.
ALEXANDRE V. BOROVIK
But the skill level required of today’s highly computerizedtechnologies cannot be adequately staffed with today’s medianskills. [The emphasis is mine—AB.]
A voice from the left (Elliot [14]), on the contrary, suggests that education hasbeen intentionally dumbed down:
We need, I should say, to look for an analysis in the direction of globaldevelopments in the capitalist labour process—especially the fragmen-tation of tasks, the externalization of knowledge (out of human heads,into computer systems, administrative systems and the like)—and theconsequent declining need, among most of the population, regarded asemployees or workers, for the kinds of skills (language skills, mathe-matical skills, problem-solving skills etc.) which used to be common inthe working class, let alone the middle classes. This analysis applies touniversities and their students. Dumbing-down is a rational—from thecapitalist point of view—reaction to these labour-process developments.No executive committee of the ruling class spends cash on a productionprocess (the production of students-with-a-diploma) that, from its pointof view, is providing luxury quality. It will continuously cut that qualityto the necessary bone. It is doing so. This, to repeat the point, is aglobal tendency rooted in the reality of capitalist production relations.
But Greenspan [18, Chapter 14] appears to take a more relaxed view on changesin economic demand for education:
While there is an upside limit to the average intellectual capabilities ofpopulation, there is no upper limit to the complexity of technology.With [. . . ] an apparently inbred upper limit to human IQ, are we des-tined to have an ever smaller share of our workforce staff our ever moresophisticated high-tech equipment and software?
Many may disagree with this claim—but it may nevertheless influence politicaland business decisions.4.
Implications for mathematics education
We have to realise that it is no longer an issue whether the role of mathematicsin society is changing: the change is being ruthlessly forced on us by Adam Smith’s‘invisible hand’.In particular, changing economic imperatives lead to the collapse of the tradi-tional pyramid of mathematics education. Let us look at the diagram in Figure 2.The diagram is not made to any scale and should be treated qualitatively, notquantitatively. The left hand side of the pyramid suggests how the distribution ofmathematical attainment looked in the mid 20th century, with pupils / students /graduate students at every level of education being selected from a larger pool ofstudents at the previous level. In the not so distant past, every stage in mathematicseducation matched the economic demand for workers with a corresponding level ofskills. From students’ point of view, every year invested in mathematics educationwas bringing them a potential (and immediately cashable) financial return.The traditional pyramid of mathematics education was stable because every levelhad its own economic justification and employment opportunities. I have includedas the Appendix the
Post Office Entrance Examination from 1897 which is beingcirculated among British mathematicians as a kind of subversive leaflet. A century
ALLING A SPADE A SPADE 7
Figure 2.
Pyramids of economic demand for mathematics educa-tion (qualitative schemes, not to scale, but higher levels of educa-tion correspond to higher levels in the pyramids).ago, good skills in practical arithmetic opened up employment opportunities forthose in the reasonably wide band of the diagram on the left, the one which hasnow become the bottleneck of the ‘hourglass’ on the right. Nowadays this level ofskills is economically redundant; its only purpose is to serve as an indication of,and as a basis for, a person’s progress to higher, more economically viable, levelsof mathematics education.The right hand side of the pyramid suggests what we should expect in the future:an hourglass shape, with intermediate levels eroded. Certain levels of mathematicseducation are not supported by immediate economic demand and serve only asan intermediate or preparatory step for further study. From an individual’s pointof view, the economic return on investment in mathematical competence is bothdelayed and less certain. Once this is realised, it seems likely to weaken the economicmotivation for further study.Many practitioners of mathematics education [13] and sociologists [15] are com-ing to the same conclusion:
Studies of the actual demands of everyday adult practices reveal thatmost occupations involve only a low level of mathematical content andexpose the disparate natures of everyday and school mathematics. [15,p. 1][. . . ] most jobs that currently require advanced technical degrees areusing that requirement simply as a filter. [13, p. 21]
The cumulative nature of learning mathematics makes a “top-heavy” model ofeducation unsustainable : what will be the motivation for students to strugglethrough the neck of the hourglass? Whether they realise it or not (most likelynot) children and their families subconsciously apply a discounted cash flow anal-ysis to the required intellectual effort and investment of time as compared to thesubsequent reward.Education (or at least state-run education) is a sector of the economy where realconsumer choice does not exist. Of course, there are a couple of choice points atwhich students and their families can decide what to study—but not how. Thereis no real choice of schools and teachers. From the economics point of view, thestate education system in England is the same as the state education in the formercommunist block (and this phrase is not intended as criticism of either of them).But it is the
Aeroflot business model of yesteryear:
ALEXANDRE V. BOROVIK
Aeroflot flight attendant:“
Would you like a dinner ?”Passenger:“
And what’s the choice ?”Flight attendant:“
Yes—or no .” In the economy of no-choice, a contributor, say, a worker or a learner, has onlyone feasible way of protecting his interests: to silently withhold part of his labour.The communist block was destroyed by a simple sentiment:
If they think they pay me let them think I am working.
Mathematics education in the West is being destroyed by a quiet thought (oreven a subconscious impulse):
If they think they teach me something useful, let them think I am learn-ing.
On so many occasions I met people who proudly told me:
I have never been good at mathematics, but I live happily without it.
They have the right to be proud and confident: they are one-man trade unions whohave withheld their learning—and, even they have won nothing, they have not beendefeated by the system.Elizabeth Truss [25] proposes a “supply-side reform” of education and skillstraining as a solution to the hourglass crisis. But supply-side stimuli work bestfor large scale manufacturers and suppliers. In mathematics education, the keylinks in the supply chain are children themselves and their families; in the global“knowledge economy” too many of them occupy a niche at best similar to that ofsubsistence farmers in global food production, at worst similar to that of refugeesliving on food donations. And supply-side economics does not work for subsistencefarmers, who, for the escape from the poverty trap, need demand for their workand their products, and demand with payment in advance —not in 15 or 20 years.Mathematics education has a 15 years long production cycle, which makes supply-side stimuli meaningless.An additional pressure on mathematics education in the West is created by thedivision of labour at an international level: in low wage economies of countries likeIndia, learning mathematics still produces economic returns for learners that aresufficiently high in relation to meagre background wages and therefore stimulateardent learning. As a result, the West is losing the ability to produce competitivelyeducated workers for mathematically intensive industries.Should we be surprised that the pyramid of mathematics education is no longera pyramid and collapses?5.
The neck of the hourglass
The mathematical content of the neck can be described in educationalist termi-nology used in England as Key Stage 3 (when pupils are aged between 11 and 14)and Key Stage 4 (when pupils are aged between 14 and 16) mathematics:
Key Stage 3 mathematics teaching [. . . ] marks a transition from themore informal approach in primary schools to the formal, more abstract mathematics of Key Stage 4 and beyond. [16, p. 6]
ALLING A SPADE A SPADE 9
It is informal concrete mathematics and more abstract formal mathematics thatmake the two bulbs of the hourglass.Why do we need abstract mathematics? A highly simplified explanation mightbegin with the fact that money, as it functions in the modern electronic world, is amathematical abstraction, and this abstraction rules the world.Of course, this always was the case. However, in 1897 competent handling ofmoney required little beyond arithmetic and the use of tables of compound interest,and clerks at the Post Office were supposed to be mathematically competent foreveryday retail finance (see Questions 7 and 8 in the Appendix). Nowadays, themathematical machinery of finance includes stochastic analysis, among other things.Worse, the mathematics behind the information technology that supports financialtransactions is also very abstract.Let us slightly scratch the touchscreen of a smartphone or tablet and look atwhat is hiding behind the ordinary spreadsheet .I prepared the following example for my response to a report from ACME
Math-ematical Needs: Mathematics in the workplace and in Higher Education [3] . Thereport provides the following case study as an important example of use of mathe-matics. The food operations controller of a catering company that supplies sand-wiches and lunches both through mobile vans and as special orders forexternal customers has developed a spreadsheet that enables the cost ofsandwiches and similar items to be calculated. [. . . ]
This task would not be too challenging to Post Office clerks of 1897, and wouldbe dealt with by ordinary arithmetic—with the important exception of the “devel-opment of a spreadsheet”. Let us look at it in more detail.
Figure 3 I used this example in my paper [7].
Anyone who ever worked with a spreadsheet of the complexity required for thesteps involved in producing sandwiches should know that the key mathematical skillneeded is an awareness of the role of brackets in arithmetical expressions and anintuitive feeling for how the brackets are manipulated, something that is sometimescalled “structural arithmetic” [16] or “pre-algebra”. At a slightly more advancedlevel working with spreadsheets requires an understanding of the concept of func-tional dependency in its algebraic aspects (frequently ignored in pre-calculus).To illustrate this point, I prepared a very simple spreadsheet in
OpenOf-fice.org Calc (it uses essentially the same interface as
Microsoft Excel ). Figure 4
Look at Figure 3: if the content of cell
C14 is SUM(C8:C13) and you copy cell
C14 into cell
D14 (see Figure 4), the content of cell
D14 becomes
SUM(D8:D13) andthus involves a change of variables. What is copied is the structure of an algebraicexpression, not even an algebraic expression itself. And of course this is not copyingthe value of this expression: notice that the value 85 becomes 130 when moved fromcell
C14 to cell
D14 !Intuitive understanding that
SUM(C8:C13) is in a sense the same as
SUM(D8:D13) can be achieved, for example, by exposing a student to a variety of algebraic prob-lems which convince him/her that a polynomial of a kind of x + 2 x + 1 is, from analgebraic point of view, the same as z + 2 z + 1, and that in a similar vein, the sum C8 + C9 + C10 + C11 + C12 + C13 is in some sense the same as
D8 + D9 + D10 + D11 + D12 + D13 .However the computer programmer (the one who does not merely use spreadsheets,but who writes background code for them), needs an understanding of what it meansfor two expressions to be “the same”. Experience suggests rather clearly that themajority of graduates from mathematics departments of British universities, as well
ALLING A SPADE A SPADE 11 as the majority of British school mathematics teachers, do not possess languagethat allows them to define what it means for two expressions in a computer codeinvolving different symbols (and, frequently, different operations) to be “actuallythe same”.This is a general rule: when a certain previously “manual” mathematical proce-dure is replaced by software, the design and coding of this software requires a muchhigher level of mathematical skills than is needed for the procedure which has beenreplaced—but from a much smaller group of workers.6.
Long division
For simplistic discussions in the media, the neck of the hourglass can be sum-marised in just two words: long division .One of my colleagues who read an early draft of this paper wrote to me: “I would not touch long division, as an example, with a ten-foot pole,because it leads to wars.”
But I am touching it exactly because it leads to wars—to the degree that thewords “long division” are used as a symbol for the socio-economic split in Englisheducation [11].Why is long division so divisive? Because it is remarkably useless in the everydaylife of 99% of people. We have to accept that the majority of the population do notneed “practical” mathematics beyond the use of a calculator, and from the “practi-cal” point of view long division can follow slide rules and logarithm tables into thedustbin of history. But why are long multiplication and long division so critical forsqueezing the learners through the hourglass neck? Because many mathematiciansand mathematics educators believe that these “formal written methods” shouldbe introduced at a relatively early stage not because of their “real life relevance”but with the aim of facilitating children’s deep interiorisation of the crucially im-portant class of recursive algorithms which will make the basis of children’s laterunderstanding of polynomial algebra—and, at later stages, “semi-numerical” algo-rithms, in the terminology of the great Donald Knuth [20]. However there is nothingexceptional about long division: many other algorithms can play in mathematicseducation the same propaedeutic role, and all of them could be similarly dismissedas not having any “real life relevance” because they are needed only by a relativelynarrow band of students, those who are expected to continue to learn mathematicsup to a more advanced stages and to work in mathematics-intensive industries. Inshort, “long division” is an exemplification of what I later in this paper call “ deepmathematics education ”.The left-wing camp in education draw a natural conclusion: long division ishard, its teaching is time- and labour-consuming and therefore expensive, and itwill eventually be useful only for a small group of high-flyers—so why bother toteach it?This is indeed the core question: I heard claims that fractions have to be excluded from the school curriculum for the samereason: only a small minority of school students will ever need them in real life.“Who of the colleagues present here have lately had to add 23 and 37 ?”—this question was asked at one of the recent meetings of experts in mathematics education.
Does the nation have to invest human and financial re-sources into pushing everyone through the hourglass neck?Or should it make a conscious effort to improve the qualityof mathematics teaching, but only for a limited number ofstudents?
This is the old conundrum of the British system of education. A recent BBCprogramme [6] has revealed that Prince Charles in the past lobbied for more (aca-demically selective) grammar schools. The former Education Secretary (Labour)David Blunkett told about his exchanges with Prince Charles:
I would explain that our policy was not to expand grammar schools, andhe didn’t like that.He was very keen that we should go back to a different era whereyoungsters had what he would have seen as the opportunity to es-cape from their background, whereas I wanted to change theirbackground . [The emphasis is mine—AB.]
This is a brilliant formulation of the dilemma, and it is especially good in thecase of mathematics education because the hourglass shape of economic demandfor different levels of mathematics education puts the emotive word “escape” on asolid economic foundation: it is the escape through the hourglass neck.While I would be delighted, and relieved, to be convinced by arguments to thecontrary, at this point I can see the solutions offered by the Left and the Right ofBritish education politics as deficient in ways that mirror each other: • The Left appear to claim that it is possible to have quality mathematicseducation for everyone. While their position is sincerely held, still, as Isee it, it leads to inconsistencies which can be avoided only by loweringthe benchmark of “quality” and ignoring the simple economic fact thatwhat they call “quality education” is neither needed by, nor required from,learners in their life, present and future, outside school. • The Right appear to claim that administrative enforcement of standardswill automatically raise the quality of education for everyone. It is also asincerely held position, but, as I see it, it leads to inconsistencies whichcan be avoided only by preparing escape routes for their potential voters’children in the form of “free schools”.My previous analysis has not made any distinction between “state” and “private”schools; this reflects my position—I do not believe that mainstream private schools,or “free schools” (even it they are privatised in the future) make any difference inthe systemic crisis of mathematics education.7.
Back to
Z¨unfte ? In relation to mathematics, social factors and, consequently, social division oflabour attain increasing importance for a simple reason: who but families are pre-pared to invest 15 years into something so increasingly specialised as mathematicseducation?What instructional system was in place before the division-of-labor sweatshopsglorified by Adam Smith? The
Zunft system . In German,
Zunft is a historic termfor a guild of master craftsmen (as opposed to trade guilds). The high level ofspecialisation of
Z¨unfte could be sustained only by hereditary membership and
ALLING A SPADE A SPADE 13 training of craftsmen, from an early age, often in a family setting. It is hard not tonotice a certain historical irony. . .The changing patterns of division of labour affect mathematics education inevery country in the world. But reactions of the government, of the educationcommunity, of parents from different social strata depend on the political and eco-nomic environment of every specific country. So far I analysed consequences foreducation policy in England; when looking overseas and beyond the anglophoneworld, one of more interesting trends is mathematics homeschooling and “mathcircles” movements in two countries so different as the USA and Russia. In bothcountries mathematically educated sections of middle class are losing confidence intheir governments’ education policies and in the competence of the mathematicseducation establishment, and are choosing to pass on their own expertise throughhomeschooling as a modern
Zunft .Some of the economic forces affecting education are brutally simple, and theprincipal barrier facing potential homeschoolers is purely financial. Mainstreameducation fulfils an important function of a storage room for children, releasingparents for salaried jobs; if parents were to spend more time with children, rates ofpay would have to be higher. A family cannot homeschool their children withoutsufficient disposable income, part of which can be re-directed and converted into“quality time” with children.Statistics of mathematics homeschooling are elusive, but what is obvious isthe highest quality of intellectual effort invested in the movement by its leadingactivists—just have a look at books [10, 23, 27]. At the didactical level, many in-ventions of mathematics homeschoolers are wonderful but intrinsically unscalableand cannot be transplanted into the existing system of mass education. I wouldsay that their approach is not a remedy for the maladies of mainstream education;on the contrary, the very existence of mathematics homeschoolers is a symptom of,and a basis for a not very optimistic prognosis for, the state of mass mathematicseducation.Still, in my opinion, no-one in the West has captured the essence of deep math-ematics education better then they have.8.
Z¨unfte and “deep mathematics education”
At the didactic level, bypassing the hourglass neck of economic demand for math-ematics means development of deep mathematics education . I would define it as
Mathematics education in which every stage, starting from pre-school, isdesigned to fit the individual cognitive profile of the child and to serveas propaedeutics of his/her even deeper study of mathematics at laterstages of education—including transition to higher level of abstractionand changes of conceptual frameworks.
To meet these aims, “deep” mathematics education should unavoidably be joined-up and cohesive. The Moscow Center for Continuous Mathematics Education, , emphasises this aspect by putting the word “continuous” into its name; it fo-cuses on bridging the gap between school and university level mathematics, while homeschoolerstend to start at the pre-school stage.
To give a small example in addition to the already discussed long division, I useanother stumbling block of the English National Curriculum: times tables. Thefollowing is a statutory requirement:
By the end of year 4, pupils should have memorised their multiplicationtables up to and including the 12 multiplication table and show precisionand fluency in their work. [22]
This requirement is much criticised for being archaic (indeed, why 12?), cruel andunnecessary. But to pass through the neck of the hourglass, children should knowby heart times tables up to 9 by 9; even more, it is very desirable that they knowby heart square numbers up to 20 = 400, because understanding and “intuitivefeel” of behaviour of quadratic functions is critically important for learning algebraand elementary calculus.The concept of “deep mathematics education” is not my invention. I borrowedthe words from Maria Droujkova, one of the leaders of mathematics homeschooling.Her understanding of this term is, first of all, deeply human and holistic.In her own words , The math we do is defined by freedom and making. We value mastery—with the understanding that different people will choose to reach differentlevels of it. The stances of freedom and making are in the company’smotto:
Make math your own, to make your own math.
When I use the word “deep” as applied to mathematics education, Iapproach it from that natural math angle. It means deep agency andautonomy of all participants, leading to deep personal and communalmeaning and significance; as a corollary, deep individualization of everyperson’s path; and deep psychological and technological tools to supportthese paths.
Droujkova uses, as an example, iterative algorithms, and her approach to thisconcept is highly relevant for the discussion of the propaedeutic role of “long divi-sion”:
From the time they are toddlers, children play with recursion and iter-ation, in the contexts where they can define their own iterating actions.For example, children design input-output “function machines” and con-nect the output back to the input. Or experiment with iterative splitting,folding, doubling, cutting with paper, modeling clay, or virtual models.Or come up with substitution and tree fractals, building several levelsof the structure by iterating an element and a transformation. Grown-ups help children notice the commonalities between these different ac-tivities, help children develop the vocabulary of recursive and iterativealgorithms, and support noticing, tweaking, remixing, and formulating ofparticular properties and patterns. As children mature, their focus shiftsfrom making and remixing individual algorithms to purposeful creationand meta-analysis of patterns. For example, at that level children cancompare and contrast recursion and iteration, or analyze information-processing aspects of why people find recursive structures beautiful, orresearch optimization of a class of recursive algorithms. Private communication.
ALLING A SPADE A SPADE 15
Maria Droujkova describes a rich and exciting learning activity. But it wouldbe impossible without full and informed support from children’s families. To bringthis education programme to life, you need a community of like-minded and well-educated parents. It could form around their children’s school (and would almostinevitably attempt to control the school), or around a “mathematical circle”, infor-mal and invisible to the educational establishment and therefore free from adminis-trative interference; or, what is much more likely in our information technology age,it could grow as an Internet-based network of local circles connected by efficientcommunications tools—and perhaps helped by parents’ networking in their profes-sional spheres. These “ communities of practice ”, as Droujkova calls them using aterm coined by Wenger [26], are
Z¨unfte at the new turn of history’s spiral. I seenothing that makes them unfeasible.I wish mathematics homeschoolers the best of luck. But their work is not arecipe for mainstream education.9. “Deep mathematics education”: Education vs. training
Who knows the difference between education and training?For those of you with daughters, would you rather havethem take sex education or sex training? Need I say more?
Dennis Rubin
The witticism above makes it clear what is expected from education as opposed totraining: the former should give a student ability to make informed and responsible decisions .The same message is contained in the apocryphal saying traditionally attributedto a President of Harvard University who allegedly said, in response to a questionon what was so special about Harvard to justify the extortionate fees, “We teach criteria.”
Let us think a bit: who needs criteria? Apparently, people who, in their lives,have to make choices and decisions. But millions of people around us are not giventhe luxury of choice.This is the old class divide that tears many education systems apart: educationis for people who expect to give orders; training is for ones who take orders. Math-ematics, as it is taught in many schools and universities, is frequently reduced totraining in a specific transferable skill: the ability to carry out meaningless repeti-tive tasks. Unfortunately, many of the students who I meet in my professional lifehave been, in my assessment, trained, not educated: they have been taught to thetest, and at the level of rudimentary procedural skills which can be described as akind of painting-by-numbers.This divide between education and training remains a forbidden theme in mathe-matics education discourse in England. But a better understanding of what makeseducation different from training would help, for example, in the assessment ofpossibilities offered by new computer-assisted and computer-based approaches tomathematics learning and teaching. I would not be surprised if computerisation ofmathematics training could be achieved easily and on the cheap—but I also thinkthat any attempt to do that is likely to be self-defeating. Indeed I believe in a basicguiding principle: if a certain mathematical skill can be taught by a computer, thisis the best proof that this skill is economically redundant—it could be best done by computers without human participation, or outsourced to a country with cheaperlabour. (For readers who remember slide rules, this is like using computers forteaching and learning the technique of slide rule calculations. By the way, you canfind on the Internet fully functional virtual slide rules, with moving bits that canbe dragged by a mouse, see Figure 5.)
Figure 5.
Figure 6.
A screen shot from an advert for
PhotoMath : pointyour smartphone at a problem in the textbook, and the answer isinstantly produced. Source: http://vimeo.com/109405701.Unfortunately, almost the entire school and a significant part of undergraduatemathematics, as it is currently taught in England, is likely to follow the slide rulesinto the dustbins of history. Figure 6 shows an advert for a smartphone app
Pho-toMath , it has gone viral and enjoys an enthusiastic welcome on the Internet.Mathematical capabilities of
PhotoMath , judging by the product website are . ALLING A SPADE A SPADE 17 still relatively modest. However, if the scanning and optical character recognitionmodules of
PhotoMath are combined with the full version of Yuri Matiasevich’s
Universal Math Solver , it will solve at once any mathematical equation orinequality, or evaluate any integral, or check convergence of any series appearing inthe British school and undergraduate mathematics. Moreover, it will produce, at alevel of detail that can be chosen by the user, a complete write-up of a solution, withall its cases, sub-cases, and necessary explanations. Figures 7 and 8 show that, un-like industrial strength software packages
Maple and
Mathematica , UniversalMath Solver faithfully follows the classical “manual” procedures of mathematicstextbooks.
Figure 7.
A screen shot from
Universal Math Solver : a fewintermediate steps of the analysis of behaviour of the function y = 11 x − xx Figure 8.
A screen shot from
Universal Math Solver : agraph of the function y = 11 x − xx which cannot be entrusted to a computer .This is, in essence, what mathematics homeschoolers are trying to develop.I am a bit more cautious about the feasibility of setting-up and developing asystem of “deep mathematics education” at a national level. It is likely to beexpensive and raises a number of uncomfortable political questions. To give justone example of a relatively benign kind: in such a system, it could be desirable tohave oral examinations in place of written ones. The reader familiar with the Britishuniversity system, for example, can easily imagine all the political complicationsthat would follow.10. “Deep mathematics education”: Phase transitions andmetamorphoses We are caterpillars of angels.
Vladimir Nabokov
I am old enough to have been taught, in my teenage years, to write computercode in physical addresses, that is, sequences of zeroes and ones, each sequencereferring to a particular memory cell in the computer. My colleague, an IT expert,told me recently that he and people who work for him passed in their lives through
ALLING A SPADE A SPADE 19 × × have their own life in a new mathematical world, much widerand richer than arithmetic—in particular, this means that “long multipli-cation” and “long division” are revisited in symbolic form;(6) interpretation of polynomials as functions; coordinates and graphs;(7) rational functions (ratios of polynomials) in two facets: as fractions revis-ited in symbolic form, and as functions;(8) and something that is not usually mentioned in school mathematics: under-standing that the behaviour of a rational function f ( x ) /g ( x ) as a function is dictated by its zeroes and poles (singularities), that is, by roots of thenumerator f ( x ) and denominator g ( x ), thus revisiting factorisation at anew level—see Figure 9 for an example;(9) and, finally, something that is not always mentioned in undergraduatecourses: the convergence radii of the power series11 + x = 1 − x + x − x + · · · Figure 9.
This problem: “Find a rational function which has agraph with vertical and oblique asymptotes as shown on this draw-ing” is a long way from the primary school fractions and ratios,but it is about ratios—this time of variable quantities. It is evenmore useless in the “everyday life” than fractions; its value liesin providing an example of a link between algebra, geometry andtopology as well as giving a tangible example of “asymptotic be-haviour”, a concept of crucial importance for many applications ofmathematics. Perhaps someone who has not mastered fractions atprimary school still has a chance to reach, in his/her later years,the level of understanding of elementary algebra and pre-calculusnecessary for solving this problem, but this is likely to be the ex-ception rather than the rule. The problem and drawing by JulianGilbey, reproduced with his kind permission.and arctan x = x − x x − x · · · equal 1 because, in the complex domain, the first of the two functions, therational (and hence analytic) function f ( z ) = 11 + z has poles z = i and z = − i , both at distance 1 from 0, and because thesecond function is an integral of the first onearctan z = (cid:90) dz z . Even ignoring the stages 8 and 9, we have six deep and difficult changes of themathematical language used and of the way of thinking about mathematical objects.Each of these six steps is challenging for the learner. But they constitute a goodpreparation for facing and overcoming future changes in professional work.I have used the classical school algebra course and a bit of calculus as an exam-ple. I accept that mathematics can be taught differently. I myself can offer somemodifications—for example, why not introduce children, somewhere after level 1,
ALLING A SPADE A SPADE 21 to a toy object-oriented programming language of the kind of
ScratchJr , and,after level 6, to some appropriately simplified version of a Haskell -kind [19] lan-guage of functional programming? But, I wish to re-iterate, I refrain from anyrecommendations, especially if they require a mass scale re-education of the armyof teachers.However, every approach to learning mathematics, if it leads to a certain levelof mastering mathematics, will inevitably involve several changes of the underlyingconceptual framework and the language of mathematical expression, at every stageincreasing the level of abstraction and the compression of information. What even-tually matters is the degree of compression (and the latter more or less correlateswith the number of phases of development through which a student passed). Manyundergraduate mathematics students come to university with a depleted ability tocompress their mathematical language further, and this is happening because theirprevious “phase transitions” were badly handled by their teachers.
The potential for further intellectual metamorphoses is themost precious gift of “deep mathematics education”.
Conclusion
I came here knowing we have some sickness in our systemof education; what I have learned is that we have a cancer!
Richard Feynman,
Surely You’re Joking, Mr. Feynman!
In this paper, I have attempted to describe how deepening specialisation and di-vision of labour in the economy affects the mathematics education system, changesits shape, undermines its stability, leads to a social split in mathematics education,and (at least in England) provokes political infighting.I wish to reiterate that I am not taking sides in these fights. I do not wish tolay blame on anyone, or criticise anyone’s views. My paper is a call for a sober,calm, and apolitical discussion of the socio-economic roots of the current crisis inmathematics education.Mathematics at the level needed for serious work, say, in electronics and infor-mation technology, requires at least 15 years of systematic stage-by-stage learning,where steps cannot be arbitrarily swapped or skipped. After all, it’s about growingneuron connections in the brain, it is a slow process. Also, it is an age-specificprocess, like learning languages.Democratic nations, if they are sufficiently wealthy, have three options:(A) Avoid limiting children’s future choices of profession, teach rich mathe-matics to every child—and invest serious money into thorough professionaleducation and development of teachers.(B) Teach proper mathematics, and from an early age, but only to a selectedminority of children. This is a much cheaper option, and it still meets therequirements of industry, defence and security sectors, etc.(C) Do not teach proper mathematics at all and depend on other countries forthe supply of technology and military protection. ScratchJr allows the learner to build iterative algorithms—see a discussion of their peda-gogical value in Droujkova’s quote above—by moving and snapping together
Lego -style blocks ona touchscreen.
Which of these options are realistic in a particular country at a given time, andwhat the choice should be, is for others to decide.I am only calling a spade a spade.
Acknowledgements
I first used the “pyramid” diagram in my talk at the Mathematics Colloquiumat the Middle East Technical University, Turkey, in April 2007, and I thank Ay¸seBerkman for her kind invitation to give that talk. The paper was developed into itspresent form thanks to my involvement with CMEP, the Cambridge MathematicsEducation Project. I thank my CMEP colleagues for many useful discussions—butneither they nor CMEP are responsible for my views expressed here.I am deeply grateful to Julia Brodsky, Dmitri Droujkov and Maria Droujkovafor generously sharing their ideas.I thank Frank Wagner for finding for me the word
Zunft ; I knew its Russiananalogues, but could not find an appropriate English term.Michael Barany, Gregory Cherlin, David Edwards, Rosemary Emmanuel, Jean-Michel Kantor, Alexander Kheyfits, Roman Kossak, David Pierce, Seb Schmoller,Victor Sirotin, and several mathematician colleagues who preferred to stay anony-mous provided useful comments on my text and/or helped to improve its style andgrammar. I am grateful to them all—but they do not bear any responsibility forwhat is said in this paper.Julian Gilbey kindly allowed to reproduce his problem and Figure 9.And I thank the anonymous referee for the most helpful advice.
References [1] J. R. Abel and R. Deitz, Job polarization in the United States: awidening gap and shrinking middle, Liberty Street Economics, 21 No-vember 2011, http://libertystreeteconomics.newyorkfed.org/2011/11/job-polarization-in-the-united-states-a-widening-gap-and-shrinking-middle.html .Accessed 11 July 2012.[2] D. Acemoglu and D. Autor, Skills, tasks and technologies: implications for employment andearnings, 10 June 2010. http://economics.mit.edu/files/5571 , accessed 11 July 2012.[3] ACME, Mathematical needs. Mathematics in the workplace and in higher education. AdvisoryCommitty on Mathematics Education (ACME), June 2011. ISBN 978-0-85403-895-4.[4] P. Anderson, Intermediate occupations and the conceptual and empirical limitations of thehourglass economy thesis, Work, employment and society 23 no. 1(2009), 169–180. DOI:10.1177/0950017008099785.[5] D. Autor, The polarization of job opportunities in the U.S. labor market: implications foremployment and earnings. Center for American Progress and The Hamilton Project, May2010. http://economics.mit.edu/files/5554 , accessed 11 July 2012.[6] BBC, Prince Charles ‘tried to influence government decisions’. 29 June 2014. .[7] A. Borovik, L’exemple de la modelisation, Commentaire 138 (2012) 500–502.[8] H. Braverman, Labor and monopoly capital: the degradation of work in the twentieth century.1974. Republished: Monthly Review Press, 1998. ISBN: 9780853459408[9] E. Brynjolfsson and A. McAfee, The second machine age: work, progress, and prosperityin a time of brilliant technologies. W. W. Norton & Company, 2014. ISBN-10: 0393239357,ISBN-13: 978-0393239355.[10] A. Burago, Mathematical circle diaries, Year 1: complete curriculum for grades 5 to 7. Amer-ican Mathematical Society, 2012. ISBN-10: 0821887459. ISBN-13: 978-0821887455.[11] J. Clifton and W. Cook, A long division: Closing the attainment gap in England’s sec-ondary schools. Institute for Public Policy Research, 2012. . ALLING A SPADE A SPADE 23 [12] T. Cowen, Average is over: oowering America beyond the age of the great stagnation. Pen-guin, 2013. ISBN-13:9780525953739. ISBN-10:0525953736.[13] D. Edwards, The math myth. The De Morgan Gazette 5 no. 3 (2014), 19–21. ISSN 2053–1451. http://education.lms.ac.uk/wp-content/uploads/2014/07/DMG_5_no_3_2014.pdf .[14] P. Elliott, a blog post on 9 July 2011 on the Times Education Supplement website, .[15] J. Gainsburg, School mathematics in work and life: what we know and how we can learnmore. Technology in Society 27 (2005) 1–22.[16] A. D. Gardiner, Teaching mathematics at secondary level. The De Morgan Gazette 6 no. 1(2014), 1–215. ISSN 2053–1451.[17] M. Goos, A. Manning, and A. Salomons, Job polarization in Europe. American EconomicReview: Papers & Proceedings 99 no. 2 (2009), 58–63. .[18] A. Greenspan, The map and the territory: risk, human nature and the future of forecasting.Allen Lane, 2013. ISBN-13: 978 0 241 00359 6.[19] G. Hutton, Programming in Haskell. Cambridge: Cambridge University Press, 2007.[20] D. E. Knuth, The art of computer programming. Vol. 2, Seminumerical algorithms. Addison-Wesley, 1981. ISBN 0-201-03822-6.[21] F. Levy and R. J. Murnane, The new division of labor. How computers are creating the nextjob market. Princeton University Press, 2004. ISBN-10: 0-691-12402-7. ISBN-13: 978-0-691-12402-5.[22] Department for Education. Statutory guidance. National curriculum in Eng-land: mathematics programmes of study. 2013. .[23] Y. McManaman, M. Droujkova, and E. Salazar, Moebius noodles: adventurous math for theplayground crowd. Delta Stream Media, 2014. [Kindle Edition, sold by Amazon.][24] E. Truss, Academic rigour and social mobility: how low income students are beingkept out of top jobs. Centre:Forum, 2011. . Accessed 10 July 2011.[25] E. Truss, The global education race. Speech at The Oxford Conference in Education,St John’s College, Oxford, 3 January 2014. , accessed 16 June 2014.[26] E. Wenger, Communities of practice: learning, meaning, and identity. Cambridge UniversityPress, 2000. ISBN-10: 0521663636. ISBN-13: 978-0521663632.[27] A. Zvonkin, Math from three to seven: the story of a mathematical circle for preschoolers.American Mathematical Society, 2011. ISBN-10: 082186873X. ISBN-13: 978-0821868737.
Appendix
Post Office Entrance ExaminationWomen And Girl Clerks
October 1897 Simplify 1 / / / / / / − / − / / / / / / / − / − / − . If 725 tons 11 cwts. 3 qrs. 17 lbs. of potatoes cost £ d. how much will 25tons 11 cwts. 3 qrs. 17 lbs. costs (sic)? Extract the square root of 331930385956. A purse contains 43 foreign coins, the value of each of which either exceeds or fallsshort of one crown by the same integral number of pence. If the whole contents of thepurse are worth £
10, 14s. 7d., find the value and number of each kind of coin. Show thatthere are two solutions. Explain on what principle you determine the order of the operations in12 + 34 ÷ − × , and express the value as a decimal fraction. Insert the brackets necessary to make theexpression mean :-Add to , divide the sum by , from the quotient subtract , andmultiply this difference by . Show that the more figures 2 there are in the fraction 0 . . . .
2, the nearer its valueis to . Find the difference in value when there are ten 2s. I purchased £
600 worth of Indian 3 per cent. stock at 120. How much Canadian 5 percent. stock at 150 must I purchase in order to gain an average interest of 3 per cent. onthe two investments (sic!)? If five men complete all but 156 yards of a certain railway embankment, and sevenmen could complete all but 50 yards of the same embankment at the same time, find thelength of the embankment. Find, to the nearest day, how long £ £ per cent. per annum (365 days) simple interest. A certain Irish village which once contained 230 inhabitants, has since lost by em-igration three-fourths of its agricultural population and also five other inhabitants. Ifthe agricultural population is now as numerous as the rest, find how the population wasoriginally divided.
E-mail address : alexandre (cid:29) at (cid:28)(cid:28)