That joke comes from the analytic calculus tradition of physics. I took a lot of physics while an undergrad, and I came out unable to do any useful physics. This is because as soon as you try even very simple problems the partial differential equations that result have either no, or extremely difficult to find solutions, and it is usually the former.
I spent years learning analytical tricks that I just don't apply.
There has been a lot of push back on this in education recently. There are Giordano's and Sherwood&Chabay's textbooks in physics now which emphasize computation over analytics. The reality is that if you want to do the vast majority of work in physics you will be doing things like reducing your problem to Ax=b, and throwing the linear algebra machinery against it.
I type this while taking a break from Strang's "Linear Algebra, Geodesy, and GPS" textbook, which I need for work right now. There is an endless need for this sort of math.
Sure, it is a matter of balance. But I flip through my diff eq books and I don't see a lot in there that I actually need. There is nothing in Strang that I don't know, and ditto for a good numerical methods book.
Even the work that Gauss did with orbits did was linear algebra and computation. He did not analytically derive the motion of Ceres, for example, but used linear algebra to compute it.
I, along with many other posters on this thread, and everyone that I have talked to, wish that there STEM education had been much more about Ax=b, the only problem we really solve, and a lot less about using huge tables of analytic integrations to try to find answers to equations that we will never encounter in the real world because they are so specially crafted as to have an analytic solution.
There are a lot of programs that are experimenting with different approaches though, and I think this will only get better with time.
I don't see any of the points you mention as particularly strong. There would be nothing wrong with presenting a more hands-on linear algebra course as a first course for freshmen. (Calling it "matrix arithmetic" does not do the notion service.) The material seems very immediate, and you can use a REPL like octave or matlab to reinforce concepts. You could deepen it as you wished by introducing more abstract concepts and by doing proofs, but I don't think that's necessary. I learned a lot from Golub and van Loan, for instance.
Second, the notion that calculus has had more impact on history/society than linear algebra is very hard to defend. Think of all the numerical modeling that has been done using matrices. Think of all the large-scale computations/optimizations that are done first by finding a linear system, or approximating one. I'm not saying there is a clear answer to this question, just that it can't be the foundation of an argument, because it's so undecidable. Another thing that makes it undecidable is cases where you need both (e.g., weather forecasting).
And saying it would be hard to teach the subject in a uniform way -- really, this is a very bureaucratic objection.
It is ideal (and indeed, not uncommon) to have linear algebra and calc 1 in parallel --- for special honors classes. It wouldn't be realistic to do it for the general freshman class. Do you realize just how many students that includes, in a big state U.? "a REPL like octave or matlab" would be wonderful, amazing, if it weren't about as realistic as the space elevator. Do you know how many endless headaches there are with software-augmented courses, even when it's just software you run in the browser? I applaud your ambitiousness but let's just say I don't want to be in your shoes when you've just told a thousand 18-year-olds to install Matlab!
>the notion that calculus has had more impact on history/society
Linear algebra is hard to even put your hands on, historically. There was no Isaac Newton figure putting the finishing touches on an amazing breakthrough "Fundamental Theorem of Linear Algebra". It is the book-keeping field of mathematics. "All the numerical modeling that has been done using matrices" almost entirely arises by discretizing differential equations so linear algebra doesn't win any points there.
The different history does not bear on the ultimate impact and general utility of the tools offered by calculus or by linear algebra.
Eigenvalues and orthogonality are not just "book-keeping". Why give forth empty slurs like this?
If you can do a decent treatment of eigenvectors in a non-honors freshman course, my hat goes off to you. Even in a 2nd-year course where students are more mature, that's hard (first time I took LA, eigenvectors were rushed at breakneck speed in the very last lecture). And while they're very important, eigenvectors are still a shadow of the historico-cultural impact of calculus.
The main use of linear algebra to me is describe local behavior of functions/dynamical systems. Sure, calculus is the setting of functions in full generality, but understanding the algebra of local interactions is an essential part of the whole -- how can you go far into calculus without understanding derivatives? Linearization is practical the way to solve 99% of engineering problems.
And I wasn't playing dumb. For one thing, even if I assumed the word cliché was used incorrectly, I still couldn't easily guess exactly what the OP really meant. Second, I don't like assuming people are using words incorrectly, it is safer to ask what is meant.
Can you substantiate "millions of transfer students between thousands of universities every quarter"? That seems like outlandish hyperbole. Irrespective of the precise number, a system with momentum is not sufficient justification for its continuation.
Linear algebra has really only come into its own since humans have had powerful computational systems at their disposal, as a large part of linear algebra's usefulness (including its relevance to the examples you describe) is efficiently organizing and performing calculations in higher order systems. That is not to say it is irrelevant or non-impactful - far from it! I'm just saying linear algebra has taken hundreds of years for its real power to be leveraged, and in the intervening time, calculus has reigned supreme as a system for modeling and describing real systems.
(FWIW, I agree with the author and the paper in question, but I think OP you replied to is absolutely correct that teaching both gives perspective that neither alone would impart.)
Studying any kind of engine (or for that matter, anything that moves), solid-state electronic device (from transistors to antennae) or power distribution network and doing any kind of design about them is literally impossible without calculus. Even algebraic approaches to these are formulated through discretization of a continuous model (which -- unsurprisingly -- you need calculus to study).
We have such a thing as databases, devices that can manipulate graphics and high-complexity economic models that are analyzed through linear algebra because Newton and Leibniz found the mathematical tools we need in order to model (continuous) change. We could mostly get along fine-ish without linear algebra, but the industrial revolution that spawned our society would be impossible without calculus, because without it you cannot design anything non-trivial that moves or runs on electricity.
Both are incredibly important. I would say that linear algebra and discrete methods are more important for computing, but calculus has had a larger impact on science especially physics. Newtonian physics - a revolution in thinking and unification of mathematics and science - is intricately tied to calculus.
Even in the example of economics that you cite, calculus is used heavily for the theoretical underpinnings for finding equilibria and deriving min/max.
I'm not. Consider all of physics. Physics is written down in terms of differential equations. You need calculus to get these equations. In terms of graphics, for example, are you magically creating only static images that never change in time? How do you think people came up with the algorithms to do things like refraction or reflection? These all involve calculus. Calculus is the most important tool in applying mathematics to the real world. It is how we derive the models for pretty much everything I can think of. Any time you want to model something where anything changes, that is calculus. Any time you want to find an optimal solutions, that is calculus. I can't think of a single area of applied mathematics that doesn't use calculus in some form. Even mathematics in general, calculus is used in some form pretty much everywhere in some form, except maybe in the foundations of mathematics like set theory or logic.
Solving them, on the other hand, is a different problem. This is where linear algebra is important. When solving an ODE or PDE, when I discretise it, all I am doing is re-writing it as a giant linear algebra problem. Quantum mechanics, for example, is dedicated to finding the eigenvalues and eigenvectors of the Hamiltonian. Linear algebra provides the tools to solve the problems posed by calculus. But it isn't always required.
In fact, if you look at the history of linear algebra, it only really became wide-spread when quantum mechanics was developing since there is a deep connection between linear algebra and quantum mechanics. Heisenberg had never heard of a matrix before despite discovering Heisenberg matrix mechanics. He was told by Max Born that what he had been doing is actually this thing mathematicians called "matrices". Now linear algebra is a required course for physicists (incidentally, linear algebra was the first lecture I ever attended at university).
So I would argue that learning calculus is the most important thing someone should know mathematically, since it is the tool we use to build models. The second most is linear algebra since it provides us the tools to solve these problems. But in my mind, knowing how to derive these models is a lot more important than being able to solve them. That and linear algebra doesn't give you any insight into why we are solving the problem that way, it is just a tool to solve the problem.
Also in Calculus 1,2,3 and so on, you learn increasingly more complicated techniques of calculus to solve more problems. But when you start numerically solving them, linear algebra doesn't care. It doesn't care if your equation came from the Einstein field equations or Newton's second law. It's all the same to it. So in a sense, I can teach you everything you'll need to know about linear algebra in a single course but the same is not true of calculus, which requires multiple courses.
I think:
Okay, then go back to pre-calculus times and somehow still invent the modern world without it and then get back to me on how irrelevant it is.
Also, you don't get to wear this shirt: http://www.topatoco.com/merchant.mvc?Screen=PROD&Store_Code=...
My feeling is that it would take a few hundred years extra (or maybe more!) to get to the point where you had a computer without calculus. Might even take a thousand. You'd only be able to do discrete stuff, so while Babbage's difference engine might be up for grabs (and mechanical calculators definitely would be) things like the Norden bomb sight or naval firing computers wouldn't be, since they're continuous not discrete systems.
You might not EVER get to computers because things like Nyquist's theorem and stuff like that doesn't even exist in a discrete-only world. Things just sometimes work and sometimes don't and it's sorta related to the frequency, but how is unclear. Aliasing is just something that randomly bites you in the ass and there's no way to develop a rigorous understanding of it and to engineer yourself out of the morass.
I would not volunteer for that mission. I like the modern world.
"The fundamental theorem of calculus is a legitimately earth-shattering breakthrough: none of the theorems in linear algebra come anywhere near it in cultural importance."
Linear Algebra has the Fundamental Theorem of Algebra (i.e. every polynomial has a solution in the complex numbers) and also the Isomorphism Theorems that say all vector spaces of n-dimensions are the same. For example polynomials of the form c+bx are equivalent to the complex numbers (over the reals) which are equivalent to two-vectors (x,y). Quantum physics is basically linear algebra, with quantum states defined as vectors in a quantum vector-space.
Don't dismiss the impact of Linear Algebra.
What do they have trouble understanding about functions?
Actually, it took me until just a few minutes ago to realize that one of the major differences between my CS-centric way of thinking and the proper mathematician's way of thinking: mathematicians treat all coextensive representations of the same object as identical, while a computer scientist tends to treat only provable intensional equality as actual identity.
Yes, it was amazingly important during all the wars and conflicts of the last century.
I don't think it was for the memory of Newton that they choose Calculus, but because it was more useful in a war machine than the alternatives.
As for the average STEM student, in my experience, most freshmen wouldn't be ready for linear algebra (unless by "linear algebra" you mean "matrix arithmetic"). They have enough trouble understanding functions in Calculus I: linear algebra's "functions are basically matrix multiplication" feature would NOT help.
Another thing to keep in mind is calculus has had much more impact on history/society than linear algebra. The fundamental theorem of calculus is a legitimately earth-shattering breakthrough: none of the theorems in linear algebra come anywhere near it in cultural importance.
Pragmatically, calculus is the most suitable mathematics for the type of testing-oriented, uniformizable/transferable pedagogy that a university really (in cold practical terms) must depend on for first-year students. It would be extremely difficult to force the kind of uniformity onto linear algebra that would be necessary if you seriously want to have millions of transfer students moving around between thousands of universities every quarter, gossiping about which linear algebra teacher gives the easiest grades, etc.