Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Featured Redesigning Mathematics Curriculum, thoughts?

  1. Aug 24, 2016 #1
    I've had a pretty poor experience going through the standardized education system in California, and now that I'm in college, I'm really fed up with how mathematics is taught (even at the college level). With this said, I thought it would be fun for me to redesign the entire math education curriculum from scratch exactly the way I would want it to be. I think this might be a fun general discussion about how math is taught in the US.

    Here is what I came up with: (The wording is a little strange, but that's because I wrote this as an answer to someone on Quora who was interested in relearning mathematics from the ground up -- something I've always wanted to do).

    1. Go online and read the short paper “The Three Crises in Mathematics: Logicism, Intuitionism and Formalism” by Ernst Snapper.
    2. Learn formal logic with: Introduction to Formal Logic by Peter Smith
    3. Go online and read the Scientific American article “Dispute over Infinity Divides Mathematicians” by Natalie Wolchover (also in Quanta Magazine)
    4. Learn Set theory with Karel Hrbacek and Thomas Jech. Introduction to Set Theory. Don’t get too bogged down with this, just enjoy the read and move on when you feel ready. Go back to it when the situation arises that you need it to move forward with math.
    5. Read a little bit about (don’t read it all the way through; just enjoy it until you get tired of it; go back to it as you work through math and see how it all fits together) Category Theory with Lawvere, Conceptual mathematics: a first introduction to categories, 2nd Edition, 2009
    6. Go online to the CSUSM Spring 2009 Math 378 course website by Prof. Aitken and download all of the class lecture notes (Ch. 0 - 10). Save them before they’re taken down, and work through these excellent notes as if they were a textbook. Learn it all as if you were given the Sports Alimak in Back to the Future series in 1985; it’s literally that good. This is the most important step in this entire list; if you do nothing else, at least do this.
    7. Get a copy of Russell’s Principles of Mathematics on amazon: https://www.amazon.com/Principles-Mathematics-Bertrand-Russell/dp/0393314049 Like the books on Set and Category thoery, read it, but not like your life depended on it. If you mastered Smith’s book on formal logic, you will master this book too, and it will help you clarify things that seem like magic in mathematics. But know, Russell’s work isn’t the end all be all.
    8. Learn about non-classical logic. Question the law of excluded middle; think for yourself — does it make sense to you? Do you believe physical reality follows this rule? All of formal mathematics from this point on, including calculus, is built on the idea that the law of excluded middle is right. In fact, even the books by Smith and Prof. Aitken, as well as all of Set Theory assume this notion. Maybe just let this question simmer in the back of your mind and continue to read about more mathematics. Don’t forget that it’s still a valid philosophical question.
    9. Read Principles of Mathematical Analysis by Walter Rudin; accompany this read with lecture notes and free online midterm exams from Stanford’s Math 19, 20 and 21 and Harvard’s Math 1a, 1b and 112. Just google the course websites and use what you can find. You should realize that Prof. Aitken’s lecture notes should make this transition seamless. After all, his notes could could well be called “Analysis of the Natural Numbers, Arithmetic and Algebra.” He even covers some real and complex number stuff, so when you see Rudin, you should be in a very, very solid position to blow this material out of the water. Do it. When you have, congratulations, you’ve probably surpassed the majority of college graduates understanding of mathematics. But don’t stop here, you need to understand more than just two dimensional mathematics after all.
    10. Read A First Course in Topology by James Munkres. This should be tons of review by this point. You should recognize things from set theory, real analysis and logic popping up everywhere. This should be an easy A, and it comes in handy as you move up to more than two dimensions.
    11. Read Abstract Algebra by Dummit and Foote. Steps 9–11 could probably be done in any order you like, or ever simultaneously. This book should build on set theory, Aitken’s lecture notes (it’s impossible to understate how good these are) and topology should seem relevant here as well. All the stuff taught in high school math is explained here with sets and axioms.
    12. Read Linear Algebra, Vector Calculus and Differential Forms, 5th edition, by Hubbard and Hubbard. Much like Rudin should have flowed seamlessly from Aitken, Hubbard and Hubbard should role off the tongue like butter to you now. You should easily grasp this material, and you should learn it because it’s important in real life. Accompany your reading with lecture videos of Math 3500/3510 by Shifrin on youtube (excellent lectures of an honors class covering multivariable math). Supplement Hubbard and Hubbard with either: 1) Linear Algebra by Levendosky, 2) Vector Calculus by Marsden and Tromba or 3) Multivariable Mathematics: Linear Algebra, Multivariable Calculus and Manifolds by Shifrin. It’s hard to say which is better. Don’t waste money buying all 3. Personally, I’d probably buy Shifrin based on his lecture videos, and also because Marsden and Tromba is on Scribd online. You can pickup Levendosky cheaply on amazon ($30 or less). If that sounds like a super good deal, buy it. It’s excellent. One last thing to add, the case could be made that Rudin will cover enough of this material to not bother with these books — that’s fair, I’d grab at least one of these just to get exposure to it though; if for no other reason than to understand physics and economics applications.
    13. Go ahead and solidify your linear algebra because it’s really important from now on. Have Linear Algebra done Right by Axler at hand and take Berkeley’s Math 110 midterm and final exams before opening Axler’s book (you can find them online easily enough). If they are easy for you, just scan the table of contents of Axler and read anything that sounds unfamiliar; skip the rest unless you want to read it. If you want to do the HW from 110 as well, then pick up a copy of Linear Algebra by Friedberg, Insel and Spence (optional). Don’t spend too much time on this. Just make sure you have Hubbard and Hubbard, and Math 3500/3510 down really well (youtube). Glaze through Axler to patch up anything not covered in Abstract Algebra and multivariable mathematics.
    14. Now you have gotten to the point where you can go online and buy any math book that interests you, and you should be able to just learn it with ease. Explore whatever you want. Algebraic topology, differential geometry, differential topology, complex analysis, physics, cryptology, computer science, statistics, anything at all. My advice, try to learn about multilinear algebra and tensors in depth. I don’t know why, but multivariable math textbooks don’t teach it, in fact the only school that I know of that teaches it is Stanford in their Math 52h class. The sky is the limit man! Have fun!

    I feel like somewhere between steps 5 and 6 there should be a lesson on linguistics and the human limitations of semantic understanding, and how this shapes our ability to understand anything that's linguistic, like mathematics. And to highlight the difference between linguistic things and the things they represent.
    Last edited by a moderator: May 8, 2017
  2. jcsd
  3. Aug 24, 2016 #2
    Nice list! First I was a bit skeptical because the list begins with a lot of logic oriented material, and category theory, which might be off-putting for some people, particularly because the applications cannot be seen just yet. But you did tone down a bit on this, saying not to get too bogged down with some of the material.

    Then it starts going to Rudin's book which is personally more familiar territory for me (I'm a theoretical physicist, not mathematician, but still really like mathematics). Well of course all those topics are very important. I am assuming this would make a pure mathematics course of course, because of the omission of applied modules. Although you could say that these are optional courses at later years. My point is for students who do care about the applications, they would of course want to learn about them in the degree, and may be impatient to wish to learn it sooner rather than later. For example dynamical systems, differential equations, statistics, variational calculus, optimization, ... not to mention subjects that relate to the sciences.

    I myself will use some of the material you stated to solidify my mathematical education, so thanks for that. Interests are algebraic topology, differential geometry, functional analysis, and in particular their applications to certain function spaces which appear in field theory.
  4. Aug 25, 2016 #3
    I would have the applied stuff come after this list in the later years as you said. This way everyone knows exactly what is going on, and can fully grasp the applications with total understanding. For example, probability theory and differential geometry can come next, followed by a high level statistics course. Diff. equations should be easy enough to grasp after doing all this. It would probably suffice to hand out supplementary lecture notes to cover the material in a short time (like a few weeks, or simply just cover the material throughout a course that makes use of differential equations, presenting the material as a special case of a more general mathematical framework students already know).

    So, for example, variation calculus can be learned within the physics class that contains its use. And more generally, regular old calculus should also be learned in physics class rather than in math class. In math class one should learn mathematics! In physics class, one should learn physics! Etc. I think this sort of curriculum would be very complementary and would produce graduates who knew their stuff extremely well.

    I would hope that emphasizing the journey of understanding pure logic before mathematics, and the interesting things this approach teaches about human cognition, linguistics and linguistic understanding, would lead to a more fulfilling learning experience for students anyway, circumventing the desire for immediate applications.
    Last edited: Aug 25, 2016
  5. Aug 25, 2016 #4
    Yes I understand this opinion, and to some extent I shared and still share it. But i'm not entirely convinced. The proper understanding of variational calculus, for instance is highly technical. I wouldn't want to deprive someone of using variational calculus without a fuller understanding, although you can argue that this is something they can do in their own time. However I certainly wouldn't want to wait until I can derive equations of motion and field equations from action principles.
    In any case at least in physics it is true that we need to be able to apply mathematics without knowing the full theory, and even researchers do that, although often they need to improve their mathematics in order to deal with technicalities. One argument in favour of 'shortcut' learning at least for physics is the lack of time.

    Now in the case of mathematics, it is a different story. But then what is meant by a math course? Nowadays it is common to do joint courses, say mathematics with economics. Also universities will have titles for the course such as Mathematics, Appllied Mathematics, Pure Mathematics. I do think the flexibility is a good thin; one is able to choose what they want. So one thing is to argue that there should exist a good number of courses like you describe. It is a much more extreme statement to say that all courses should be like that. I don't think you would get much support for the extreme statement.

    Now whether you solution is the best way to learn mathematics (even if one wishes to apply it in the future), i don't know. Personally only experience will tell, if I compare the two approaches in the future, when i'm more acquainted with both, I will be able to judge better.

    It may be interesting to look up the calculus trap.
  6. Aug 25, 2016 #5
    I've never heard of the calculus trap until now, but it sounds very relatable!

    I also agree that sometimes there just isn't enough time to do both math and, for example, engineering, or a science like physics, and learn them both really well, which is really unfortunate because then I think you get people who go on to be physicists or engineers without really understanding the mathematics too well, and this might lead them astray as they get more and more advanced; or worse, in might lead their intuition to funky places that just wouldn't make sense if they learned the math right from the begining.

    As far as curriculum goes, and implementation, I agree that this might be a little impractical for everyday use, but at most schools something like this isn't even an option (like the calculus trap explains), and that's just too bad because I think it would serve some people really well.

    Now, on the other hand, even if it's extreme, why not introduce formal logic in elementary school? Set and category theory too. I mean, category theory could easily be taught in a music class when teaching music theory (which is an extremely cool way to learn music). And set theory could be taught as an extension of logic, then students could be weaned off of constant problem solving and learn a little about the peano axioms, and what mathematical proof is all about, as well as learn that it actually is the foundation of everything that anyone who uses any math does with or without their knowledge of that fact.

    Philosophy electives in high school can take this foundation and run further, teaching epistemological notions of consciousness for example. Metaphysics is also a great place to go from a school system like this, students shouldn't have their grades punished for these classes or anything, nor should they be required, but to have them just to encourage free thinking, and to bring back class discussion and a focus on the socratic method rather than standardized testing would be nice in my opinion.


    But anyway, if you're actually interested in trying to learn the stuff from this list and see how it goes I'd be curious to hear about the results!
  7. Aug 25, 2016 #6
    Actually I think introducing logic, some further set theory and category theory in schools is a great idea. Again one needs to tone it down a bit, so one needs new material invented, new books, ... For example the first time i looked at category theory it was offputting because many examples I did not know at the time. Now I know some more. So bearing in mind the lack of mathematical experience, it is certainly a challange to create such material, but not impossible.

    One argument one may have against this is saying that some students might never use this in their life. Well you could question this assumption itself, but another counterargument is that there are many things students learn in high school that they do no directly use in their life (although it may have an indirect influence) anyway, in other subjects. Moreover learning some logic could be benefitial for critical thinking, which is prety much applied everywhere. The International Baccalaureate, which I did, is a program that has a compulsory part called Theory of Knowledge, which does include logic and philosophy of mathematics. But it would be nice to also see logic in a maths class. Oh yes and as you said, philosophy can take this even further.

    I do think that in school the mathematical education needs to change a lot. Personally, here in the UK, I see so many unmotivated students of maths. I believe this is because of how it is taught. At least students should know that mathematics is not just about a tool. Some people may go through high school without awareness of the existance of pure mathematics. People are not too keen in becoming calculators, they want to use their creative power.

    Yes, I can tell you the results of my learning, although it may take some time. I'm quite busy at the moment.
  8. Aug 27, 2016 #7


    Staff: Mentor

    How so? I went through high school and some college in California, albeit lots of years ago, but it served me well. What exactly is the problem you're trying to fix?

    The first 6 or 7 items on your list in the OP have to do with logic and set theory. The so-called "New Math" of the 60s also had these areas as priorities, and were taught with mixed success, to be charitable, as many of the teachers in public schools did not have strong enough backgrounds in mathematics to understand this stuff, let alone teach it.

    Are you asserting that calculus isn't mathematics? Many of the applications of calculus are drawn from physics, but the underlying concepts are mathematical in nature.

    Much of this seems very impractical to me, especially such esoteric subjects and epistemological notions of consciousness and metaphysics. A major problem at too many U.S. high schools is that students are graduating and are unable to do simple algebra or even grade-school arithmetic. I don't see teaching them metaphysics as being a viable alternative.

    I agree that some knowledge of logic is necessary for students to be successful with proofs in geometry, and much later, linear algebra and beyond, but I don't see how human cognition and linguistics tie into or are important in a mathematics curriculum.
  9. Aug 27, 2016 #8
    How do you think this issue can be resolved?
  10. Aug 27, 2016 #9
    First of all, the logic in geometry is, in fact, logic. It's no different from algebra.

    Now, the idea is that the people who struggle with basic algebra and arithmetic are struggling for a reason, and I don't think it's systematically teachers' fault, or students' lack of effort, rather I believe that the curriculum is to blame. Maybe most of the students who struggle with basic algebra struggle because they can't see the logical framework underlying the algebra. Is it such a radical idea to teach abstract algebra to the students who don't seem to be accepting the notion that they just have to believe their teacher without question? Why not teach them group theory after logic? Show them that functions are the mapping between two sets, the domain and codomain. Ordered pairs are the cross product of two sets; ordered pairs are usually things that are just casually tossed around in high school without ever explaining what they are. Right? "They're just the X and Y coordinates" ... "Look, let's draw a cartesian plane, see, the ordered pair is a point on the plane -- rise over run." But what they don't tell you is that the real numbers are a field, and that the cartesian plane is the set R^2. Thus the cartesian plane is the set of ordered pairs formed by R cross multiplied with R. This is why a set of basis vectors spans the space, since the set of all linearly independent basis vectors can take on any ordered pair (in dim = 2). This is why dimension is defined as the cardinality of the set of basis vectors that span the space -- Ie, dim 2 = ordered pairs, dim 3 = ordered triples, dim 4 = ordered quadruples. This is why the number of components in a vector corresponds with dimension, an element of the basis of R^3 is a vector of the form [x, y, z] -- this is no coincidence. This is why dimension is often written as R^1, R^2, R^3 and R^4, since R^4 is R x R x R x R, therefore R^n is R x R ... n times. And if you think of a line as a set of points and the set of real numbers as the "real number line (which is the set of points in R)," then R^n represents a set of lines -- a set of real number lines, or the cardinality of the set of basis vectors of a space -- since the span of a vector goes off in either direction like a line. In other words, R^n represents the coordinate axis, and it's why R^2 is drawn with two real number lines. This is why the cartesian plane is justified and drawn like a cross. And in fact, the cartesian plane in two dimensions is {(x, y) | x,y e R}, and R^2 = {a[x] + a[y] | a e R}.

    Why do we delude our students? Why can't we just tell them the truth so it's not surprising where the rules and properties in problem solving come from?

    I believe that regular curriculum can be taught as special cases of the truth of mathematics (Gasp, what a novel idea!), and the philosophy classes are there to look at justifications for the axioms that are used without justification in mathematics -- of course they are assumptions, but to students who don't accept algebra, I can assume they'll also ask, say, "how can you be sure of the axiom of infinity?" Or, "How does it work in real life if there is no explanation for the axiom of infinity?" These questions are clearly metaphysical, so why not educate people on metaphysics? It seems like a plenty worthy subject to understand.

    Finally, I do believe that Calculus is more like physics than mathematics. Calculus is to math is like what oranges are to orange juice in a cup. There's nothing interesting about calculus, you just take the concept of dividing by a number close to zero -- big deal. You find areas by stacking a bunch of rectangles next to each other and add them all up -- big deal. These are things that are needed primarily for physics; I mean, Newton believed there was no difference between math and physics when he created calculus. If you read his "Princpia Mathematica" you'll see that it was all based on euclidian geometry, and the idea that geometry was reality, and that reality was geometry. Therefore, by studying geometry he was studying physics, and by inventing calculus the intention was to explain the reality of physical objects. He was making the philosophical assumption that geometry is metaphysically true. Now, it seems like there are people who are in extreme disagreement over whether or not Mr. Newton was right about his metaphysical assumptions. In fact, nobody even knows today. As an aside, isn't this kind of embarrassing for academics? People don't even know what mathematics really is; ie, created vs discovered argument that has no answer because people actually do not know the answer. The approach I suggest is the logicist's approach, because to me it seems that the laws of logic are more fundamental and obvious than anything else. Unfortunately for me, 1) the law of excluded middle is questionable (though all mathematics assumes it's metaphysically true), and 2) most professional mathematicians either believe, like newton, that physical reality is simply math, or they somehow believe that math is just a set of rules, and apparently the content of these rules don't matter. Either way, right now in today's world mathematics seems to be working, and right now everything is based on a set of rules, which is obviously ZFC. The point is to teach students about ZFC and why everything is currently the way it is in the field of mathematics, because there are no other reasons. I don't understand why students are left in the dark here.

    The problem of educators not being educated enough to teach what needs to be taught is an alarming problem, but it all has to start somewhere. It doesn't make sense to constantly delude students by lying to them year after year as the story changes in mathematics with things like "oh, last year we just told you X because it's easier to understand. We actually lied to you, it's really Y." After a while of this it's no surprise that kids grow up hating math, and not learning it.

    I think my solution is better.
    Last edited: Aug 27, 2016
  11. Aug 27, 2016 #10


    Staff: Mentor

    I should say that my meaning was some students graduate without being able to do simple arithmetic. There are also significant numbers of them who are functionally illiterate. I base this on my years in a community college, and the large number of remedial classes in algebra and below and English.
    I'm not sure that it can be solved, short of significant changes in (U.S.) society. My wife is a school psychologist, and she reports that there are many parents who don't place any value on education. Our schools here (U.S.) don't provide a track for jobs in the trades, and the students who don't do well don't have the intention or aptitude to succeed in college.
  12. Aug 27, 2016 #11


    Staff: Mentor

    I disagree. In algebra there are a number of properties and laws, such as the addition property of equations, and the laws of exponents, to name a couple. To my mind, these are quite different from and less sophisticated than such concepts of proofs as contrapositives or contradictions.
    Based on what evidence?
    Yes, that is a radical idea. Good teachers will give an explanation for why some property or law works the way it does.
    And how would knowledge of the properties of a field help them solve an equation like 3x + 7 = 16, something that many of them struggle with?
    To paraphrase Jack Nicholson in "A Few Good Men": "They can't handle the truth!"
    What exactlyl is the "axiom of infinity?" I have a bachelor's and a master's, both in mathematics, and this is not a term I've ever heard. I disagree strongly that metaphysics is something that needs to be taught or that philosophy be taught as a prerequisite to mathematics. I should warn you that discussions of philosophy aren't permitted at this site.
    The rate of change of a function and the area under a curve are purely mathematical. As I said before, the physics comes when you attach these concepts to some application.
    The concept of the limit, which is used in the definitions of both the derivative and the definite integral, wasn't crystallized until long after Newton (and Leibniz) did their work. These are purely mathematical concepts.
    It's well known that physicists and other scientists use mathematics as a tool. What you're saying sounds to me like you're equating the hammer a carpenter uses with the stairs that he builds.
    I would say this as, "He was making the philosophical assumption that geometry is metaphysically true grounded in reality."
    What we do know is that a lot of Newton's writings were on mysticism, which are irrlevant to mathematicians generally.
    We don't add the word "metaphysically."
    And in my opinion, much of what you propose is completely unworkable.
  13. Aug 27, 2016 #12
    Well, this is the axiom of infinity: https://en.wikipedia.org/wiki/Axiom_of_infinity It's what justifies mathematical induction by asserting, axiomatically, that at least one infinite set exists. In other words, it's impossible to know. We can only count finitely, and so, it requires faith. Constructivist mathematicians reject this axiom and carry on without it, this means any proofs that rely on mathematicall induction become invalid. The difference between constructivist math and regular math is a metaphysical distinction, in particular, the axiom of infinity. I say students deserve to know that this controversy exists, and they deserve to know why, think for themselves, and come to their own conclusions. Hence, metaphysics is something good to teach in high school.

    As for how abstract algebra helps a student find "3x + 7 = 16," for a start the definition of a field justifies addition and multiplication, for which subtraction and division can then be derived. Here is the wiki on fields: https://en.wikipedia.org/wiki/Field_(mathematics)
    Last edited: Aug 27, 2016
  14. Aug 27, 2016 #13


    Staff: Mentor

    Any comments on the other questions I asked?
    Students are typically exposed to problems like this one in the 8th or 9th grades (it was 9th grade algebra for me, back in the late 50s). And your intention is to teach them about the field axioms before they solve problems like this? Good luck with that...
  15. Aug 27, 2016 #14
    This is exactly the intention. I don't see why it has to be any different from what we do today. We would just introduce sets early, and teach basic logic like "if a then b" stuff early as well. Then teach them about the rules of sets, then introduce group theory. That way when one approaches regular algebra there is no need for confusion. Set theory is a branch of pure logic, it shouldn't be hard to grasp because there's really not much going on in set theory. It's very general.

    As for the calculus stuff, well, analysis is mathematical, but computational calculus classes that don't teach anything but following the rules to get a numerical answer is not really mathematics to me. In fact, I'd argue that it actually isn't math, it's whatever the math is being used for in the particular computational problem being solved. If it's a physics question, then it's physics. If there is an economics question on a HW, then the student is doing economics, not math. If it's a statistical question, then are doing statistics, not math. Anyway, I'd say just teach real analysis instead of "calculus for engineers, mathematics and scientists" as it's normally done, because "calculus for engineers" implies that people don't want to understand math; and it never even tries to teach for understanding.

    Moreover, I feel like today people's understanding of math is artificial. There's just something wrong with that.

    People think that physics, economics and statistics is math, but it isn't. But all they've been doing in all their supposed "math" classes all this time actually hasn't been math, it's been a smorgasbord of assorted questions in other fields.

    Okay, but there's nothing wrong with physics and the likes. In fact, I love physics, but I don't understand why math has to be the slave to the sciences and engineers at the expense of understanding. If we taught math in math class and economics in economics class (in terms of the general math theory, again, what a novel concept) I believe our education would be greatly improved. Eventually, since we'll start teaching math in math class, the economists will be able to teach economics in terms of the general framework. This way the students can focus on the core concepts, of, say, economics and spend little time applying math to it. Likewise, in math class students don't get bogged down with economics questions.

    The curriculum should not be biased in this way as it currently is. Math should stand on its own and its classes should teach mathematics.


    Here is an example of a HW assignment that can be done in the 8th grade: "Suppose that a and b are two elements of a field F.
    Using only the axioms for a field, prove the following:
    – ∀a ∈ F, 0a = 0.
    – If ab = 0, then either a or b must be 0.
    – The additive inverse of a is unique."

    This is totally feasible.
    Last edited: Aug 27, 2016
  16. Aug 27, 2016 #15
    That is a good point. Probably the greater part of the problem is a social issue. I do think teachers should try to motivate the students as much as they can, be there is only so much they can do. Part of the responsability will lie with the parents.
  17. Aug 27, 2016 #16
    For the record, this is a separate, but very important issue. I totally agree with this.
  18. Aug 27, 2016 #17
    Ok I agree that basic logic is important and I personally never found it too difficult. It is in fact introduced often in say, books on analysis, because of the need for proofs.

    Well that is a bit a generalization. Often it is the teachers fault, and/or the student lack of effort. You can certainly ask what are the causes of these issues, and I think they are sociological, cultural and psychological in nature, for the most part.

    It depends on what you mean by logical framework. If you mean what I think you mean, i.e. learning the required subjects for the construction of the real numbers as in done in analysis, even before one can solve equations, I think is not effective. I have two main reasons for this.

    Firstly there is a logical framework in doing algebraic manipulations. There are some rules, and by following these rules one learns how to manipulate abstract symbols. A computer indeed does the same thing, and just because I said we don't like being computers, doesn't mean computation isn't a very significant part of mathematics. And when we are doing manipulations, we are doing proofs too, so I think this is what needs to be emphasized.

    The field axioms for real numbers is certainly known to students, even though they don't know it is called 'field axioms' or that mathematicians may think of a multitutde of examples of fields. They are the basic rules which are taught early on about algebraic manipulations. I do think the teacher should point out the name 'field', and maybe even give a definition. This is tricky however, because of the important fact that not many examples of fields are known to the student at this point, so it can be confusing.

    Secondly, logicians also do manipulations. They also manipulate symbols in logic. There are also rules of logic they are following. They also compute, even if what they compute may be more abstract. Even logic has computation.

    It is easier to learn how to use a set of rules to do manipulations with objects that are more concrete. Hence why solving equations is taught first. A number 5.4 is certainly more concrete than a mathematical proposition.

    So I'm in favour of teaching basic logic simultaneously with solving equations, so that the connection between proofs and the algebraic manipulations can be seen. However I am not in favour of teaching logic and set theory strictly before teaching how to solve quadratic equations, because of the lack of concrete examples.

    Actually this is a point I have been mentioning in these posts: the examples. We learn much more, and in a better way, by knowing examples and doing exercises. Well of course you should know this (actually do you? How long have you been in college for?). This is not logic's fault. It just so happens that are our brains have evolved in that way. Also, not only we remember concepts that we learnt by example and exercise, we also remember what has been learnt visually. There are many concepts I can remember in topology and analysis precisely because I have a visual picture of them. I usually remember a definition after I remember the picture or concept. That is why I remember a lot, not because I have memorized. Actually things i do not remember tend to be correlated with things I have not done many exercises in.

    So if you are going to design a class where it is not even posible to come up with examples because the students simply do not know them, you will be in trouble.
    Last edited: Aug 27, 2016
  19. Aug 27, 2016 #18
    Can you really say you can do computations in calculus if you have not practised? Sure, we learn how to explain those computations more precisely after courses in analysis, but I feel like ommiting calculus is wasteful, since even historically analysis was motivated by calculus in the first place. Also, if you know some analysis\calculus you should know computations in calculus are not as easy and trivial as you described. They can be extremely difficult. The rigorous logic behind the subject serves to make the problems of calculation more visible and clear, but they do not necessarely make it simpler.

    That actually gives me an opportunity to say that historically both the development of physics and mathematics are highly interconnected. While mathematics provided great tools to solve physical problems, physics provided great motivation to develop further mathematics. Yes, Newton did invent calculus as a tool, because he wanted to calculate the gravitational pull of the moon by the Earth and the mathematics of his day was not good enough to solve this problem. I think you should delve more deeply into the history of mathematics and you will see what I mean.

    Recently string theory has motivated and even produced a ton of new mathematics. Many calculations mathematicians were struggling with were solved by string theorists even before they could catch up - and yes the string theorists did invent new mathematics. In fact, one of the fathers of string theory, Edward Witten, was the first (and so far the only) physicist to be awarded the fields medal.

    When I go into research I want to communicate with mathematicians, even though I am a theoretical physicist. I have strong faith in this interplay which historically has been so successfull. It is inefficient for us to isolate ourselves from each other, when so very often our problems are very similar in nature, even though the language describing the problems and the motivation behind doing the problems might be different.
    Last edited: Aug 27, 2016
  20. Aug 27, 2016 #19
    I do like how you bring up the interconnectedness of math and physics. I guess I have to concede to this point. I also know that the calculations can be very difficult.

    As a theoretical physicist, do you ever think it's possible to know the explanation before the result? I think the root of what I'm saying involves what might be a fact that understanding probably is always historically forced to come after doing. Since it's impossible to explain before experiencing. Maybe math is taught in this same spirit, that first you must accidentally, or mysteriously, figure out something and then you must go back to figure out why -- where as I'd rather try to determine what's logically possible before trying anything. And then, once all possibilities are known through the laws of logic itself, then anything that happens is explainable. But you know, this is probably never how it can be. So I'd understand it if those who designed our curriculum wanted to train people for real work in the field, that they'd feel that students have to get used to trying first and explaining later. So this might be the actual reason for the current way math is taught.
  21. Aug 27, 2016 #20
    I think it is possible to have some insights about the explanation, but one must work hard to have the full explanation, if the problem is difficult. I also thinking there are layers to understanding. When Newton discovered calculus, he did not suddenly know everything there is to know about calculus. It was a basic level of understanding which was improved and deepened as time went by. So now we understand much better, thanks to the work of many individuals. And by the way, it is conceivable, that maybe we will never have the 'full' knowledge of a certain subject.

    Yes you are probably right that the way mathematics is taught mimics the historical discoveries, to some extent. That is why we learn things that have been discovered in the 17th century first, before learning what was discovered in the 20th century, and this trend is not only true in mathematics.

    The idea of determining what is logically possible before trying anything is excellent, and this is why it is so widely used nowadays. Even physicists use it. Particle physics is based on some constraints (like Lorentz invariance) which determine what is possible and what is not possible. We try to logically derive what is impossible given the assumptions of our theory, but sometimes it is very hard to be rigorous. This is certainly one good reason to communicate with mathematicians.

    So even the problem of determining what is logically possible is far from trivial, and the technicalities highly depend on what you are studying, but this approach is very fruitful in research. That doesn't mean that research focused on calculation is also not fruitful - it is.

    And yes, real work requires good problem solving skills which can only be acquired by practise and being able to solve complex problems with the tools you have at your disposal - and not falling for the calculus trap. Of course one day, someone comes along, and decides to invent a new tool to solve a problem, which is much more effective than the tool used by peers, and everyone is amazed!
    Last edited: Aug 27, 2016
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?