Musings on the physicists/mathematicians barrier

[Doodle Bob]

All very well, but this discussion is in the context of differential forms being a replacement for vector calculus for ordinary physicists and engineers. As per my original point, I believe froms to be unsuited to this task. Whether by design or immaturity, they are not a suitable topic of study for most physicists involved in the study of either electromagnetism and especially fluid dymanics.

This may very well be true. This topic might indeed be a bit abstruse for the average physics undergrad. However, one should keep in mind that a good century and a half ago, the very same thing could have been said about the relationship between linear transformations and matrices. There was at the time not much use for them among the physicists until quantum physics came around.

I must say, though, having just recently read The Large Scale Structure of Space-Time by Ellis and Hawking that a good knowledge of forms (and other elements of differential geometry) are essential to the understanding of GR.

If I remember correctly, nrqed's inital post was in the context of several other threads on the topic of differential forms and possibly topology, where the supposed benefits of forms were being lauded to nrqed who, quite rightly, simply didn't see the benefit in the frankly massive amount of formalism required to study these topics. He's absolutely right. Topology in paticular is now a disaster area for the newcomer. 100+ years of invesigations, disproofs, counter examples, theorems and revisions have lead to the axioms and definitions of topology being completely unparsable.

A great many topology books offer nothing but syntax with no sematics at all. Differential forms texts fare little better. To a good physicist, sematics is everything, and hence the subject will appear to the great majority of them to be devoid of use. That's actually a problem with a lot of mathematics, and modern mathematics in paticular. Syntax is presented, but sematics is frequently absent.

I've always considered it very bad manners to criticize someone else's discipline as worthless, and the above seems to me very bad manners.

Modern mathematics is indeed very complex, and a very wild field to start out in. However, formalism and logic is the mortar that keeps it all together. Without proofs and rigorous thinking, math is just magic. Hence, a great deal of research seems to be more fancy window-dressing than anything substantial. But, every so often a big theorem comes into view: I'm thinking of two within my mathematical career: the Fermat-Wiles Theorem and the solving of the Poincare conjecture (and hence of Thurston's Geometrization Conjecture).

These results probably don't mean to you (they are after all worthless to electromagnetism) but they mean a great deal to me and other mathematicians.

 

[AKG]

Through axioms! You define d/dx to be an operator that:
(1) is a continuous operator
(2) satisfies (d/dx)(f+g) = df/dx + dg/dx
(3) satisfies (d/dx)(fg) = f dg/dx + df/dx g
(4) satisfies dx/dx = 1

and I think that's all you need.

Why like that? Why not:

\frac{d}{dx} : f \mapsto \left( x \mapsto \lim _{h \to 0}\frac{f(x+h) - f(x)}{h}\right )

 

[ObsessiveMathsFreak]

I've always considered it very bad manners to criticize someone else's discipline as worthless, and the above seems to me very bad manners.

I was speaking from an andragogical standpoint.

 

[Doodle Bob]

I was speaking from an andragogical standpoint.

that may be the case. but i do not see any mention of "adult learners" in this post or any of the others.

 

[ObsessiveMathsFreak]

that may be the case. but i do not see any mention of "adult learners" in this post or any of the others.

I think it's safe to say not many children would be learning differential geometry from textbooks.

 

[Doodle Bob]

I think it's safe to say not many children would be learning differential geometry from textbooks.

Well, that's a very slippery way of avoiding the essence of my assertion: you spend a great deal of time knocking modern mathematics (and topology in particular) as insignificant technobabble and very little talking about curriculum and undergraduate pedagogy.

 

[mathwonk]

as my 8th grade teacher used to say about our reaction to the class clown: "you're only encouraging him."

 

[Chris Hillman]

The shortcomings of a technical education: whom is to blame?

Oh dear, I wrote a long reply to OMF, then belatedly noticed a crucial remark:

I was speaking from an andragogical standpoint.

Sigh... Oh well, here's the longish post I wrote predicated on the (mistaken?) assumption that OMF is a twenty-something recent college graduate:

I've read a lot of books on differential forms. Not that one, but still many others.

I take it that one of them was Spivak's book, Calculus on Manifolds? You do realize that the goal of this book was not intended to do what you ask? I will go out on a limb and guess (from your username and the context of this thread) that your undergrad major was math, not physics or engineering. If so, I wonder if you might not have been in the wrong major.

Many of which purport to have applications to physical sciences, but usually just throw down the differential forms version of Maxwell's equations by diktat with little or nothing in the way of semantics. Worked examples are few, probably for the reason that the worked out question would be longer than the route taken by regular vector calculus.

Well, if a worked example was the first thing you wanted, it is certainly too bad that you didn't start with the book by Flanders...

I feel the main impetus for differential forms was to formalise something that was never really valid in the first place, namely concepts like; df or equations like
df = \frac{\partial f}{\partial x} dx + \frac{\partial f}{\partial y} dy
instead of the actual equation
\frac{df}{dt} = \frac{\partial f}{\partial x} \frac{dx}{dt} + \frac{\partial f}{\partial y} \frac{dy}{dt}

1. Well, I guess this depends upon what you mean by "valid". Is a "linear approximation" invalid simply because it is not an identity?

2. Trust me. While historians of mathematics have apparently not yet tackled the career of Elie Cartan (despite his extraordinary influence on the development of modern mathematics), I probably know more about his interests than you do. In particular, I know something about his interests in Lie algebras, differential equations and general relativity, as well as integration.

For Cartan's work on the central problem in Riemannian geometry (in fact a whole class of problems involving differential equations), try Peter J. Olver, Equivalence, Invariants, and Symmetry, Cambridge University Press, 1995. Notice that this work lies at the heart of the Karlhede algorithm in gtr. For more about Cartan's involvement in the early development of gtr, see Elie Cartan-Albert Einstein : letters on absolute parallelism, 1929-1932. English translation by Jules Leroy and Jim Ritter ; edited by Robert Debever, Princeton University Press, 1979. For more about Cartanian geometry (common generalization of Riemannian and Kleinian geometry), try R. W. Sharpe, Differential geometry : Cartan's generalization of Klein's Erlangen program, Springer, 1997. For "Newtonian spacetime", see the chapter in Misner, Thorne, and Wheeler, Gravitation, Freeman 1973.

It is, or IMO should be, very striking that these sources are almost completely independent of each other. Cartan's work is characterized by a remarkable coherence of purpose and scope, yet adds up to so much that even whole commitees of authors can attempt to explain only bits and pieces.

For an attempted overview of Cartan's influence on modern mathematics, Francophones can try Elie Cartan et les mathématiques d'aujourd'hui, Lyon, 25-29 juin 1984 : the mathematical heritage of Elie Cartan, Société mathématique de France, 1985. For anglophones, an important textbook on mathematical physics, which is contemporary with Cartan's career, which emphasizes the utility of differential forms, and which might provide a few hints about why these techniques should be mastered by any serious student of mathematics, is Courant and Hilbert, Methoden der mathematischen Physik. This book went through various German language editions beginning in 1924. It has been translated into English (Interscience Publishers, 1953-62), and IMO remains valuable to this day!

I have decided for myself. I don't approve of differential forms. At least, not as a replacement or improvement for vector calculus. That's just my own opinion, but I would ask others to consider this point of view before imposing forms arbitrarily on undergraduate courses.

Gosh. You certainly seem to be embittered. That is especially unfortunate since this really is such a lovely subject.

About your experience in school, I'd just comment that I think it is very unfair to assume that faculty make arbitrary decisions when designing curricula. I have spent enough time as a math student (and teacher) that I think I can confidently assure you that decisions of this kind, while never easy, are not made lightly.

this discussion is in the context of differential forms being a replacement for vector calculus for ordinary physicists and engineers. As per my original point, I believe froms to be unsuited to this task. Whether by design or immaturity, they are not a suitable topic of study for most physicists involved in the study of either electromagnetism and especially fluid dymanics. They may, like other advanced mathematical topics, be of use in describing new theories or methods, but this thread is about their promotion for more basic studies, as per nrqed's initial post.

Well, I happened to know two of the mathematicians (John Hubbard and Beverly West) who redesigned the undergraduate curriculum at Cornell two decades ago, and I know that they did not take this responsibility lightly! And decades later, I see that the math courses have been redesigned again (good, these decisions should be revisited at least five times per century), but differential forms remain firmly at the heart of the applied mathematics background for the engineering major. See http://www.engineering.cornell.edu/programs/undergraduate-education/minors/applied-mathematics.cfm:
and note these two courses:
MATH 321 Manifolds and Differential Forms II
MATH 420 Differential Equations and Dynamical Systems

Quite frankly, I feel that this demanding curriculum is one reason why the Cornell Engineering School is one of the best: it ensures that graduates have mastered the techniques they will need to work as engineers (or to go on to graduate work in engineering).

If I remember correctly, nrqed's inital post was in the context of several other threads on the topic of differential forms and possibly topology, where the supposed benefits of forms were being lauded to nrqed who, quite rightly, simply didn't see the benefit in the frankly massive amount of formalism required to study these topics. He's absolutely right. Topology in paticular is now a disaster area for the newcomer. 100+ years of invesigations, disproofs, counter examples, theorems and revisions have lead to the axioms and definitions of topology being completely unparsable.

For those whose minds are not made up, I would offer an alternative take on the question of why math courses are so demanding. New math builds on old math. New ideas which rest upon old ideas are not neccessarily any harder to learn, as long as the student masters the older context first. A mathematically trained intuition is a very different thing from what a random process (natural selection) has equipped most humans with. Humans are adapted to learn, and do so very well, and many humans are probably quite capable of retraining their intuition to the point of being able to apply powerful theories like topology and the theory of manifolds in applications in physics, engineering, and other areas. But this retraining takes time.

Unfortunately, larger social issues force universities to try to churn out their graduates in four years, rather than the six to ten years which in my view would be more reasonable for most undergraduate students. This is really a problem too big for the universities, but I feel that it would be more intelligent to adjust upwards both the standard age when an educated youngish person is expected to enter the workforce, and the standard age when an oldish person is expected to retire.

A great many topology books offer nothing but syntax with no sematics at all. Differential forms texts fare little better. To a good physicist, sematics is everything, and hence the subject will appear to the great majority of them to be devoid of use. That's actually a problem with a lot of mathematics, and modern mathematics in paticular. Syntax is presented, but sematics is frequently absent.

I think that if you accept what I said just above, it may be that our positions are not so different after all. Perhaps our real difference is over whether you should blame the math faculty at your school, or the politicians who consistently fail to tackle important long range social issues in the country where you were (mis?)-educated.

 

[Hurkyl]

To a good physicist, sematics is everything

That's wrong, of course. Without syntax, you are incapable of doing calculations, or communicating with others.

And besides, one can define sematics for a formal system in terms of the syntax itself, so you can't say that any formalism is inherently devoid of semantics.

But that's not the main reason I'm responding...

A great many topology books offer nothing but syntax with no sematics at all.

I'm going to have to call you on this one. Most of the terms one would learn in elementary topology can evoke an immediate geometric picture: open set, closed set, compact set, interior, exterior, boundary, compact set, connected set, path, pathwise-connected, sequence, sequentially compact, continuous function... Feynman even tells a story how his (mathematical) colleagues would come to him and describe whatever scenario they had been working, and Feynman would generally build a mental picture of what they're describing, and would was generally quite accurate at guessing the result of their analysis.

Of course, one of the strengths of the axiomatic method is that it is syntactic, allowing the reader to interpret it in whatever context he desires. I guess, though, that causes a problem for a reader uninterested in forming those interpretations for himself, despite demanding they exist.

 

[ObsessiveMathsFreak]

Well, that's a very slippery way of avoiding the essence of my assertion: you spend a great deal of time knocking modern mathematics (and topology in particular) as insignificant technobabble and very little talking about curriculum and undergraduate pedagogy.

I was not knocking modern mathematics. I was knocking the way modern mathematics is being taught. It is of course a problem throught mathematics, where we have many people who are very good at it, but very few people who are very good at teaching it.

Sigh... Oh well, here's the longish post I wrote predicated on the (mistaken?) assumption that OMF is a twenty-something recent college graduate:

Just on this and the previous post. There's no reason to assume from my statement about andragogy that I am not a recent graduate. Nor is it correct to referr to the teaching of undergraduates as pedagogy. Everything from undergraduate up is andragogy. You are teaching people who are adults, and who only learn topics they feel are relavent to them.

I take it that one of them was Spivak's book, Calculus on Manifolds? You do realize that the goal of this book was not intended to do what you ask? I will go out on a limb and guess (from your username and the context of this thread) that your undergrad major was math, not physics or engineering. If so, I wonder if you might not have been in the wrong major.

Spivak's book was by far the best book on forms that I read. By far the best book on calculus for that matter. I consider it as having completed a lot of things left out or paper over in my calculus education so far. Spivak was good because he was what so many other author were not. That is, precise. He fully explained, in the required mathematical detail, what a form was, what it did, etc, etc. He did fall down a bit on tensors, but I think without the physics behind them, tensors remain too up in the air for full conceptual understanding.

My undergraduate degree was in applied mathematics, and I consider myself an applied mathematician. I see mathematics as a disipline to be learned, studied and indeed advanced in the context of problems, be they from physics, chemistry, statistics, or even philosophical questions. In retrospect I see my degree choice as being a very good one over physics, engineering or even theoretical physics.

Well, if a worked example was the first thing you wanted, it is certainly too bad that you didn't start with the book by Flanders...

I've taken your recommendation and ordered it. If I see forms working well on a problem, perhaps I'll see them in a new light. But I must mention that I have seen them at work on a good many problems, and I have not yet seen any great advantage in the method.

Gosh. You certainly seem to be embittered. That is especially unfortunate since this really is such a lovely subject.

I would describe differential forms as many things. Formal certainly. Interesting there is no doubt. They can even be useful when one moves into higher dimensions. But lovely is not a word I would use for a topic that allows for old ghost of maths class past like dx + dy to rise up and walk the Earth once more.

Unfortunately, larger social issues force universities to try to churn out their graduates in four years, rather than the six to ten years which in my view would be more reasonable for most undergraduate students.

I would strongly think otherwise. Four years is quite a reasonable enough amount of time to spend in any undergraduate degree. Anything more would be far too much.

I understand that in the United States, when people finish their degree, they go on to do six years of coursework to obtain a Phd! I would strongly disagree with this. This is far too much to ask anyone to do. Where I am, the regieme is that Phd's are granted through research. Your research could be, usually, between three and five years. In that time, you truly do learn the skills of your trade, and I can personally say I learned at lot faster, and a lot more by researching than I ever did taking classes.

Most topics would only really require a good solid week in a workshop anyway. Differential forms for example. I spent about a month dipping in and out of it. To be honest I don't think a huge amount more is required in most fields, especially if you may not end up using the topic much. Not just differential forms, any topic. I don't agree with spending ten years in classes. I think you learn more out of them than in, on your own initiative of course.

I think that if you accept what I said just above, it may be that our positions are not so different after all. Perhaps our real difference is over whether you should blame the math faculty at your school, or the politicians who consistently fail to tackle important long range social issues in the country where you were (mis?)-educated.

I blame the mathematicians. They're not precise enough.

That's wrong, of course. Without syntax, you are incapable of doing calculations, or communicating with others.

And besides, one can define sematics for a formal system in terms of the syntax itself, so you can't say that any formalism is inherently devoid of semantics.

I think syntax and sematics should come together. In synergy. One cannot understand one without the other. I a big believer in introducing every new mathematical theory or concept via a problem, because that is invariably where it originated.

I'm going to have to call you on this one. Most of the terms one would learn in elementary topology can evoke an immediate geometric picture: open set, closed set, compact set, interior, exterior, boundary, compact set, connected set, path, pathwise-connected, sequence, sequentially compact, continuous function...

The terms might evoke intuitive ideas and pictures. The definitions certainly do not. Despite any impressions I may be giving off, I still consider myself a mathematician, and preciseness and exactness are important to me.

In this regard, even though the concept of an open set is perhaps intuative, I need a precise and clear definition to move on. Most topology books, in fact every topology book I have ever read, fails to meet this criterea. While the definitions are probably precise, they are as far from clear and intuative as it is possible to be.

For quite a while, I took a compact set to be a single point or element, as the definition given was; "A set is compact if every open cover has a finite sub-cover". Seeing this in the context of the support of the delta distibution, I took the definition straightforwardly as decribing a point, as the author had used the strict subset notation when describing a subcover. The author has sacraficed clarity for terseness, unneccesarily in my opinion. Simply stating "A set is compact if every open cover is either finite or has a finite sub-cover", would be a perfectly clear definition where of course the compactness of paticluar sets could be inferred immediately without invoking subcovers, and where notational laxity would not cause problems later down the road.

This is only one of the many examples where topology books resemble more a house of mirrors than what they should resemble, which is "Calculus on Manifolds". Definitions, examples and exercises. Explanations wouldn't go amiss either.

 

[laserblue]

You are not alone. See THE VECTOR CALCULUS GAP: Mathematics not= Physics
by Tevian Dray and Corinne A. Manogue (24 September 1998)
(http://www.math.oregonstate.edu/bridge/papers/calculus.pdf)

BRIDGING THE VECTOR (CALCULUS) GAP by TEVIAN DRAY and CORINNE A. MANOGUE
(http://www.physics.orst.edu/bridge/papers/pathways.pdf)

Bridging the Vector Calculus Gap Workshop (http://www.math.oregonstate.edu/bridge/)

Physicists and Mathematicians often use mathematics in different ways and have different viewpoints on what some mathematical symbol means and how it is to be interpreted. E.g. Mathematicians think of vectors in terms of tuples of numbers in a linear space while physicists think of something with a magnitude and a direction. Sometimes physicists create their own mathematical objects to better understand something and occasionally they muddy the water by splicing two different formalisms together such as in the use of the Pauli matrices in quantum mechanics.