Why Are Physics Classes Full of Old Stuff?

Everybody and their siblings have been linking to this Minute Physics video, an "open letter" to President Obama complaining about the way that most high school and even intro college physics classes don't teach anything remotely modern:

I'm not entirely sure where the date of 1865 comes from, but it's true, the standard intro physics sequence doesn't really touch what's normally called "modern physics," a term which is itself laughably out of date, as it generally refers to special relativity and quantum mechanics as it stood around 1935. We don't teach really new stuff until about the 300 level in college courses (junior/senior year for students on the normal track), with the possible exception of hand-wavey non-majors courses.

So, why is that, anyway? Are physics teachers and professors just totally oblivious to how backwards and out-of date their curriculum is? No, of course not, especially at the college level. We're acutely aware that what we teach is mostly old physics-- it would be difficult not to notice that we're not teaching anything remotely related to the research we do (in most cases). We're pretty much stuck teaching the material that we do, though, because of constraints that are external to physics departments.

A factoid that I picked up at an AAPT workshop about ten years ago is that only about 3% of students taking introductory college physics ever take another physics course. Now, some of that is attributable to the fact that intro courses are often dry and boring, and we should absolutely fix that. But that tiny retention rate is in large part because the students who are taking intro physics are only taking it because it's required for some other major. At Union, where I teach, we teach close to 140 students intro Newtonian physics every year, the vast majority of whom intend to major in engineering, with a smattering of chemists and mathematicians mixed in. While the class serves as the entry point to the physics major, in academic parlance it's really a "service" course-- something we're doing for another department. The next biggest chunk of our enrollments, 70-ish students per year, is the Physics for Life Sciences course, which most of the pre-med students take to prepare for the Physics section of the MCAT.

The fact that these courses are service courses first and foremost constrains what we can teach. And much as we might wish it were otherwise, the engineering and chemistry departments don't particular want us to teach the cool modern stuff. They want us to teach old physics from 1865, because that serves as the foundation for some of their courses. We have to teach classical mechanics first because that's what the departments that provide most of our students want us to teach.

High school physics ends up covering the same basic material as intro college physics for the same reason that high school biology and chemistry resemble introductory college biology and chemistry-- because that's what high school classes do. They offer classes that cover the most basic stuff for those who will never take another science class, and provide a foundation for those who go on to take another course in college.

Why do chemistry and biology teach more modern material than physics does? Because chemistry and biology as sciences developed more recently than physics did. Lord Rutherford's division of science into Physics and Stamp Collecting was snide, but not without some truth in 1900 or so when he said it. The existence of atoms as real physical things wasn't definitively settled until the early 20th century, and a solid understanding of how and why atoms combine into molecules the way they do didn't come along until quantum mechanics was worked out in the 1930's. The theory of evolution, without which nothing in biology makes any sense, wasn't put forth until the 1850's, and most of the biochemistry of life wasn't figured out until the 20th century. Chemistry and biology classes don't spend a whole lot of time on chemistry and biology from before 1865 because most of what was known about those subjects before 1865 has been superseded or cast into a completely different light by more recent discoveries.

Physics before 1865, on the other hand, had accomplished a hell of a lot that's still useful today. The basic laws of mechanics date from the late 1600's, and still work brilliantly for describing the motion of macroscopic objects moving at everyday speeds. Maxwell's equations, which might be the source of the 1865 date, provide a complete and correct description of classical electromagnetism, full stop. They're even compatible with relativity-- in fact, relativity grew out of attempts to reconcile Maxwell's equations with the rest of physics. Classical thermodynamics, the laws of which are another possible source of the 1865 date, works extremely well for describing the flow of heat in macroscopic systems (and, in fact, thermodynamics probably accounts for most of the pre-1865 material taught in chemistry classes).

We spend a lot of time teaching students about physics from before 1865 because physics from before 1865 is pretty damn useful. It's the same reason why the Math department spends most of their time teaching students about math that was developed before 1865 (differential and integral calculus, Euclidian geometry, trigonometry)-- because "old" math is still extremely useful. Blaming math and physics for their early success is kind of ridiculous, given that the stuff works. It's also an essential foundation for the cool modern stuff-- it's almost impossible to really understand quantum physics without first knowing a good deal about classical physics.

Yes, but what about Carl Sagan and Neil DeGrasse Tyson and Richard Feynman? Look, I love what they do, but Sagan and Tyson aren't even in the same business as most people teaching physics. Feynman's the only one of those three who attempts to teach people how to solve problems. As my angry quantum prof in grad school put it, though, while reading the Feynman Lectures may make you feel like you understand everything, "when you try to solve a problem, you realize that, well, that you're not Feynman."

Sagan and Tyson, and Feynman to a large extent, are popularizers, not educators. Their job is to get people fired up about science in a more general way, not to teach them how to do anything with their knowledge. Those are very different businesses-- believe me, I know, having written two popular books about cool modern physics. And while I made every effort to ensure that How to Teach Physics to Your Dog and How to teach Relativity to Your Dog are rigorous and correct (to the point of probably limiting their sales...), I wouldn't begin to claim that they teach people how to do physics. If you want to know that, you need to take actual physics classes at the college level, and you need to start with classical mechanics and E&M.

Finally, as much as I love modern physics, I have a problem with the suggestion that "old" physics is intrinsically boring. True, Newton's Laws don't fire the imagination the way Schrödinger's Cat does, but that doesn't mean you can't do cool things with classical physics. If you don't think classical physics can be cool, you're clearly not reading enough Dot Physics. Rhett's blog is one of the most consistently awesome things on the Internet, and he almost never talks about physics developed after 1865.

This is where there's a glimmer of hope regarding the complaints in the video. The problem isn't that old physics is intrinsically boring, but that the traditional way of teaching physics is kind of dull. But if you look at the stuff that Rhett, and Frank Noschese, and John Burk, and Kelly O'Shea, and the whole Global Physics Department crowd are doing, you'll see that there's a lot of room to teach "old" physics in new ways that make it much more interesting and appealing, while still satisfying our obligations to other programs and departments.

Of course, there's another way to look at this, too, which is to try to do cool things with modern physics on a conceptual level even earlier-- the whole Physics First approach to high school science. That's a cool idea, and I'd be happy to see it pushed more strongly because I think it has a lot of potential. As it is, though, it runs up against a lot of entrenched interests, so I don't give it great odds.

But anyway, if you're wondering why it is that we teach all this old stuff in physics classes, those are my slightly rant-y thoughts about why. It's not because we haven't thought about it, believe me.

More like this

Interesting points, both about physics and math. I would mention that we are in a similar situation with statistics at the college level, although my sense is that it is changing more quickly than the areas you cite.

Introductory classes in statistics are mired in the least-squares/normal distribution for data approach, with robust methods, even simple nonparametric methods, covered as an afterthought if time permits.

Sometimes I think that the old line "the military is always ready to fight the previous war" can be modified to "the sciences and math faculty are always ready to teach old material".

One thing that occurred to me after I wrote this yesterday, but couldn't find a graceful way to work in, is that the "old stuff" in physics is not entirely old. The language and mathematical apparatus that we use in discussing old physics is often of more recent vintage than the physics itself-- Maxwell wouldn't necessarily recognize his own equations written in modern form. So there is some parallel in physics to the situation in other sciences-- post-1865 math puts pre-1865 physics in a much different light than it was originally, in the same way that knowledge of molecular genetics puts biology from before 1900 in a much different light.

The other day I had a long conversation with my 6-year-old daughter about gravity. It eventually got into Einstein and general relativity in a very qualitative way (because one of the things that sparked it was that she found my copy of Weinberg's "Gravitation and Cosmology").

But the thing that really fired her imagination was my explanation of Newton's law of universal gravitation, and in particular, the idea that it's not just the Earth pulling on the apple or the Sun pulling on the Earth, it's everything pulling on everything. "You mean that sign is pulling on that tree? I'm pulling on this book? That car is pulling on our car?" She went on and on and on.

By Matthew McIrvin (not verified) on 15 Nov 2012 #permalink

I don't completely buy the service course argument because here in the UK we don't have any physics service courses. Physics majors are taught separately from day one and the physics necessary for other subjects is usually taught within the home department. However, we still pretty much have exactly the same curriculum as the in US, the only difference being that we cover some subjects that are graduate courses in the US due to the narrower focus of our degrees. I think the same could be said for most other European countries.

One of the ways I think we can introduce more modern physics into the first year of a physics degree is to introduce short courses that cover the conceptual basics. We have done this in the UK for special relativity for many years and it it about time we started doing it for quantum theory as well. In fact, spurred by the quantum informationish approach of the Schumacher/Westmoreland textbook, I know that some UK lecturers have been developing a curriculum for such a course within the Institute of Physics. There is a danger that some physicists will see this as a quantum information takeover, but the main point is that you can do some cool stuff with qubits before students have seen vibrations and waves, differential equations, etc. and impart a basic understanding of the measurement postulates in this simpler arena where students are not trying to follow how to solve the hydrogen atom at the same time.

I do agree with you that the way that physics is taught is more important than teaching the latest stuff, particularly for intro courses in classical mechanics. However, I think we can combine that with bringing down a few select topics from modern physics to earlier in the curriculum. Those topics should be chosen so that they can have some technical meat on them, e.g. basic finite dimensional QM with matrices is good but a particle physics course in which students are taught to name the different particles and count lepton numbers with no math other than arithmetic is bad.

By Matt Leifer (not verified) on 15 Nov 2012 #permalink

How could one even think of working with the cool stuff if he does not have a strong foundation? There is no use talking about Lagrangian Mechanics if you don't understand Newtonian Mechanics and if that is old then how will one proceed?The stuff taught in our first year of study ( in India) at college is also dry and boring but that does not make it trivial that it must be scraped off or replaced! I have read somewhere that a pianist must practice long boring scales for hours before he dreams of composing good music. The same thing is valid here! Dry and boring old stuff is important for understanding the new and cool stuff!

By Subhodeep Sarkar (not verified) on 15 Nov 2012 #permalink

I think the service course thing is a significant problem in that it restricts the development of alternative approaches and texts. There have been attempts to make curricula that take a more modern approach, such as Holbrow et al.'s Modern Introductory Physics, but they haven't gotten much traction, in part because they don't fit quite as neatly into the service-course needs of large departments in the US, which account for most of the textbook market. Even reformed intro curricula like the Six Ideas series and Matter and Interactions are obviously constrained by the need to hit a set of topics that more or less match the traditional engineering physics track.

I agree that intro QM with matrices would be a really cool way to bring some modern physics in earlier. I think that doesn't get done very often because for some reason Linear Algebra is regarded as more advanced math than vector calculus. I think this is probably another distorting effect of service courses, this time on the Math side-- engineers all need to take calculus, so that's the basic course, so linear algebra courses are pitched more at math majors, and go heavy on proving abstract theorems and the like. Which makes the subject more challenging than it might be with a more practical approach-- I took linear algebra in the math department as an undergrad, but it never made any sense to me until I got the swashbuckling physicist version in grad school QM and particularly Bill Phillips's Atomic Physics class.

It's a long trip from the intuition of Aristotle to Physics. The "old stuff" helps student get grounded in a history of unification (Newton unified the physics of terrestrial motion, thought to be linear, with celestial motion, thought to be circular. Maxwell unified electric phenomena with magnetic phenomena), with quantitative results and managing uncertainty, with the math of algebra (Physics isn't about memorizing equations but rather application of model; Algebra isn't about "x" but rather about the equals sign), calculus, multivariable calculus, functional calculus, probability distributions, statistics and logic and to develop a language of shared jargon so that the definition of "particle" or "energy" or "velocity" depends on the physical model being used at the time.

It is entirely possible to write out General Relativity as 10 coupled polynomial equations in elements of g (the metric), T (the energy-momentum-pressure tensor), and partial derivatives of elements of g. Handing that out to a student who has just mastered partial derivatives will not do them a favor. Some will develop a delusion that they "know GR" and some will make sign errors and claim that they have "proved the Schwarzschild metric is not a vacuum solution." With the language of differential geometry, tensors and Bianchi identities still years away, it is doubtful any student will become motivated by such a handout. And without Lie algebras, a study can only receive a pale shadow of Noether's theorem.

It's similar in biology. Darwin and Mendel are still useful, and necessary to understand where things are today.

By The Phytophactor (not verified) on 15 Nov 2012 #permalink

Prof. Edwin Taylor has suggested that teaching physics beginning with the principle of least action rather than with Newton's laws would both foster a deeper understanding and enable students to comprehend the way that physical laws are currently understood by physicists. http://www.eftaylor.com/leastaction.html#calltoaction

By Patrick M. Dennis (not verified) on 15 Nov 2012 #permalink

I suspect that conservativism (in the sense of being reluctant to change something that appears to be working) is a major factor as well. Newtonian mechanics is an excellent approximation under the conditions that most people taking introductory physics (engineering and geophysics majors) are likely to encounter in their careers. It only breaks down under certain conditions: (1) velocities a noticeable fraction of c (special relativity); (2) objects sufficiently small that their de Broglie wavelength has to be accounted for (quantum mechanics); and (3) conditions where the metric tensor is significantly different from the free space value (g00 = -1, g11 = g22 = g33 = 1, off diagonal terms vanish) (general relativity). The first of these is something non-physicists almost never encounter, and there are branches of physics where the issue seldom arises. The last arises for people who need to correct GPS signals for frame dragging, but is otherwise limited to certain branches of physics.

Item (2) is a different story. Chemists and materials scientists, as well as some electrical engineers, encounter situations where quantum mechanical effects are important. But this is a sufficiently small fraction of the people enrolling in freshman physics that most people find it easier to ignore these people at the freshman physics level, and teach them quantum in an upper-level undergraduate course within their own department. (At some universities the relevant course may even be graduate level, but the chemistry and EE departments at my undergraduate alma mater had undergraduate level courses in quantum mechanics.)

Of course, if you did decide to add a little QM in freshman physics, you would probably have to take something out, and listen to the screams of whatever department whose oxen you gore in the process. Semester/quarter scheduling constraints, if nothing else, dictate this outcome.

By Eric Lund (not verified) on 15 Nov 2012 #permalink

Eric Lund: To the best of my knowledge, QM of some sort is a staple of all chemistry undergraduate curricula. The quality of the course will vary and may embarrass physicists, but it's a requirement. I, as a chemistry instructor, would love for students to see some QM in the physics pre-reqs for chemists.

This topic ("modernizing" intro courses) comes up for chemists all the time, especially from biologists whose students complain about their chemistry requirements. From all of my department's discussions, I have two thoughts:

1) Chad, I think there's one other issue that spans both chemistry and physics. In theory, we as college educators (especially at a small lib arts school) are supposed to be able to take ANY freshman student and accept them into the curriculum and have them graduate in your major in four years (again, in theory). We cannot assume that a student has had a competent chemistry or physics course (or any course at all) coming in so the intro course acts as a leveling experience for the varied background of students. At a small school we cannot have a huge variety of intro courses for all different levels of background.

2) To me, part of this argument for modernizing first year curricula (in chemistry at least) smacks of "edu-tainment" (putting the onus on the professor to keep all students interested in every topic). We chemists cannot and should not tailor our courses entirely to the interests of biologists and premed - even if they are the majority of the students - the same way the math department doesn't tailor linear algebra to the physicists and chemists in the course. At some point students have to accept the responsibility of learning a topic without direct context to their field, so that they will be able integrate it when the need arises.

so linear algebra courses are pitched more at math majors, and go heavy on proving abstract theorems and the like. Which makes the subject more challenging than it might be with a more practical approach– I took linear algebra in the math department as an undergrad, but it never made any sense to me until I got the swashbuckling physicist version in grad school QM and particularly Bill Phillips’s Atomic Physics class.

This matches my experience exactly. I took linear algebra in the math department as a sophomore, but had no sense of how incredibly useful it is until I encountered it in graduate QM.

By AcademicLurker (not verified) on 15 Nov 2012 #permalink

I like your take on this. I thought when I saw it that the MinutePhysics take didn't think through enough how modern physics can be delivered, although as always it was brilliant in itself.

I suppose things might change when one can buy a Bell-violation apparatus with USB connections and accompanying signal analysis software for $200.

By Peter Morgan (not verified) on 15 Nov 2012 #permalink

I taught high-school physics for 30 years (all levels from Conceptual to AP). For the first several years I wanted a physics course that would whet one's appetite. I wanted my students to think physics was so cool that they couldn't wait to get to college and learn some more physics!
We did all the standard stuff, did horrible things to my truck, did the bed of nails, etc. I saved up quantum physics for the end of the year. I was having a blast, and many of my students went into technical fields or into science careers.
When I had to cram the AP curriculum into my 180 day schedule I had less fun, the kids were more stressed and I am not so sure about my success.
Damn NCLB and Race to the Top!

If the university would allow it why not schedule for the first year physics majors a 1 credit seminar course where various faculty members talk about the research they are doing? I pushed for this back in the late 1960s but got nowhere. Low and behold at CalTech they had the 1 credit seminars that one took to hear current developments in the geologic sciences (geology club it was called at the time). This did include visitors, but then the average department could include them in such a course. This gets the major a chance to see what is happening, but yet allows the presentation in somewhat the way it happened historically. All be it that spending till christmas in high school physics based upon Modern Physics learning about simple machines seems a bit much. Of course the book also dealt with how circuits with vacuum tubes work which would not be very useful today.

Chad, I respectfully disagree. You said: "High school physics ends up covering the same basic material as intro college physics for the same reason that high school biology and chemistry resemble introductory college biology and chemistry– because that’s what high school classes do. They offer classes that cover the most basic stuff for those who will never take another science class, and provide a foundation for those who go on to take another course in college."

If you are teaching students who will never take another physics class, then focussing solely on Newtonian mechanics and slowly building up to E and M is a bad choice, and it is no wonder that in APS sponsored surveys the public does not see any benefit of scientific research in their everyday life.

If they are not going to take another science class again - then discuss how the electrical juice comes out of the wall socket. There is a very broad range of physics topics (some post 1865!) that you can legitimately cover this way, and the information will be relevant for those students who do not plan on being scientists or engineers, but will be citizens and voters after graduation.

Or you can discuss QM and applications (lasers, transistors, MRI, etc.) at a qualitative level. You do not need matrix mechanics to learn the basics of solid state physics - and believe me, telling High School students the physics that enabled their cell phones and iPads, they will better appreciate that scientific research can have an influence on their lives.

By James Kakalios (not verified) on 15 Nov 2012 #permalink

Chad,

I don't agree quite agree with you here. For those of us with some background in physics, it is definitely beautiful and exciting that we can explain such a large breadth of phenomena with a handful of rules. To us, it makes sense that you'd want to invest a lot of time in understanding the subtleties of how to use these rules. I love Rhett's blog and yours as well, but I am a physics teacher, physics grad student, and generally a physics geek. I should not be the intended audience of an undergrad physics course.

Most students do not think like this. Many see physics as some kind of elaborate mathematical game that you can use to predict real-world quantities. They get that it is powerful and effective, and they get that it is hard. They just don't see why they should care, or get excited about it. And I think that it is an incredible challenge for teachers to convey this excitement with the standard first semester physics coverage of newton's laws, conservation laws, kinematics, etc.

In the course I teach we use modern pedagogical methods like active learning, investigative learning, and we focus on model building and application. My students have told me that they are very happy with the quality of learning in our recitations and labs, and we try in many ways to relate physics to real-world experiences. But just last week one of the best students told me that she dislikes physics, and hasn't liked it since high school. One of the students on the teaching staff, an undergraduate who is now a learning assistant, told me that she can't imagine how and why someone would do research in physics. This is coming from people who have excelled in the course material, and it is by no means a minority point of view.

The thing is, we are ridiculously fortunate in physics that we have modern, understandable material that is guaranteed to blow anyone's mind. We've got ideas like the twin paradox, the double slit experiment (with one particle at a time), bell's theorem, the uncertainty principle, the breakdown of simultaneity, gravitational lensing, event horizons - this stuff is totally mind warping. And it just crushes me that we let students lose interest in physics and think that it's dry and boring, without introducing them to some of these truly fascinating ideas. When I think back to why I got interested in physics, it is because I read popular science books on these subjects, so I knew that there was good stuff to come if I just fought through my fair share of inclined planes and pulleys. I'm not denying that classical physics is fascinating, but this is a point of view that comes with some mathematical maturity, in my opinion. The better your grasp of mathematics, the more deeply these concepts connect with each other. But the vast majority of undegraduate physics courses at my institution do not even assume calculus as a pre-requisite. Mechanics is not going to look sexy to these kids. And my fear is that we are losing students' interest, by not introducing them to ideas that they are guaranteed to find cool. They're not going to wait to get to the good stuff, because they don't even know that it exists. Furthermore, we are boring many of them (particularly the smartest kids) by repeating a lot of material that they've already seen in high school. If this high school physics experience was not pleasant, teachers have an added mental block they have to struggle to undo.

In my undergrad program at Swarthmore College, the first physics class I ever took was called 'the character of physical law' and it didn't have a single pendulum or an inclined plane. It was an introduction to the ideas of relativity and quantum mechanics, and it had the effect of significantly increasing the college's physics enrollment. It was a beautiful, inspired class that succinctly conveyed the truly big ideas in physics, so that you'd be motivated to chug through the next few classes in mechanics and electromagnetism. And I think it was an idea that worked. I know many people who came in wanting to major in engineering, computer science, or math, and were 'won over' to physics over by that single class. I also had lab partners who were English literature or studio art majors.

So count me in as someone who believes that we should blow minds first, and build solid foundations afterwards. It's not enough that our students are required to take our classes, we also need to earn their interest. And I would hope that doing things this way might help lift the number of students who take a post-intro physics class to more than just 3 percent.

Regards,
Aatish
@aatishb

By Aatish Bhatia (not verified) on 15 Nov 2012 #permalink

Amen James. Why not calculate the escape speed for a black hole? Why not do some Bohr model? QM is a way of looking at the world and Id rather students have a broader view than be able to calculate ( and often not well) the moment of inertia of a flywheel

When talking about waves, sneak in deBroglie. Cold atoms have smaller RMS speed and can interfer and diffract. You can describe energy bands via splitting of energy for a sin vs cos wave in an periodic potential from ionic cores Estimate the mass of a meson using a Yukawa meson

You can show them new physics without crying they dont know QFT. Show them the outer layer of the onion. Not drill down to the core of classical mechanics. Even more important is to teach them how scientists think, how science is done.

As far as teaching you can do it well with lecture/response and badly with clickers. Its about the communication skills of the teacher and class size to the first significant figure or two

Chemists see orbitals and how to make atoms and molecules early. We roll balls down inclined planes

The 1 hr supplement for (mainly) majors that does detailed demos and lots of lab tours has worked well for us for many years. A number of folks leave after the first year, but we usually we have a net gain after the first year. Most egrs take it the first year and we show them all the cool toys

I agree that you could do a more qualitative "intro to cool stuff" physics class at a lower level-- in fact, we do offer such a course, aimed at prospective physics majors, which they take in their first term. Five faculty do two weeks each (we're on a ten-week trimester calendar) on a topic related to their research. When I do it, I talk about laser cooling, one of my colleagues talks about quantum information, another does particle physics, another dark matter, another black holes, etc. It works reasonably well at getting students fired up, though the reception is more mixed than you might expect-- some of them, including some of our better students, find the frequent drastic shifts of topic kind of overwhelming.

That sort of course doesn't really meet our service obligations, though. In fact, we have a persistent problem with prospective engineers signing up for it over the summer, then immediately being advised to drop it by their advisors in engineering departments.

It's also something we can get away with doing only because we can rely on students having heard a lot of physics terminology from the old stuff in their high school classes. I can talk to them about laser cooling because I have some confidence that they've heard "momentum" and "energy" defined before, and have a hazy idea how those things are supposed to work. That provides just enough grounding for the weird stuff to make a tiny bit of sense. Take away that background prep on boring old stuff and it's much harder to talk about cool new stuff in anything but the haziest fashion.

It's easy to say that we should do more cool stuff early, but when you start to dig into the details of how to do it, it's really hard. Calculate the escape speed of a black hole? Fine, but how is that any different than calculating the escape speed of a rocket leaving the Earth using boring old classical physics? For which you need to explain Newtonian gravity, and force, and energy, and the relationships between those things. Talk about the Bohr model? Sure, but if you want it to make sense, they need to have some idea about what energy is, and why it's weird for energy to be quantized. If you want to discuss the history, you need them to know something about angular momentum as well.

Without a bunch of old stuff in the background, you end up with not much more than formula soup, with no real unifying intuition behind it. If anything, it's probably even more scattered than what we have coming out of the intro classes now.

Believe me, we've tried this. We used to do a version of the intro course that included a week or so of relativity in the classical mechanics course, and a couple of weeks of quantum in the E&M course. It sorta-kinda worked, bot only sorta-kinda. That's part of why we switched to the Matter and Interactions curriculum, which mixes in a smaller amount of modern physics-- they start with the relativistic definitions of energy and momentum, for example-- in a more coherent framework. Which sorta-kinda works. Maybe.

You could, of course, imagine doing something more "new" and conceptually oriented at the high school level, as with "Physics First." If you're going to put it at the end of the curriculum, though, that runs into a huge problem with a deeply entrenched notion of what high school science is, to say nothing of creating a bunch of problems for college courses that now rely on students recognizing some intro physics material from high school classes in order to make calculus make more sense, and so on.

It's really easy to talk about ways to show more cool stuff to students early-- I've done a lot of it over the years. When you get down to the nitty-gritty details of how to make that work , though, it's a really difficult problem. Which is one of the reasons it hasn't been fixed-- again, it's not like we're not aware of the issue. We know what's wrong, we just don't have a very good solution.

I have a few bones to pick with many of the comments here.

First, and most importantly, mechanics *is* the cool stuff too. And mechanics is relevant to everybody as well. Familiarity is not the same as "boring". And there are deep philosphical issues involved in the fact that many aspects of Newtonian mechanics are not intuitive. There is the universality of the laws of physics - that gravitation works on planets and apples. There is the fact that there are universal truths hidden underneath the most superficial layer that we see and that we can extract these from careful experimentation (objects in motion stay in motion, etc.). There is the fact that specifying the position and velocity of a particle determines its trajectory (Newton's second and third laws).

Secondly, the tendancy to superficially introduce advanced material can be very problematic. My impression is that physics is hard for some students because they are not sufficiently proficient in basic math skills. The very idea that one could confuse voltage and speed because they are both represented by v's suggests a basic failure in mathematical understanding from way back. I have also heard that the bounds on an integral are "constant", that a paragraph of words cannot be math, and a complete failure to understand dimensions (if I double the radius of a ball, how much does the volume grow by?). I don't see how talking about relativity or quantum mechanics before these basic ideas are introduced won't just encourage students to think of science as "magic", and who feels like understanding "magic" is really something they can do?

Chad,

Thanks for the compliment. I just want to ask one question: Why is it that no one ever gets on to English faculty for teaching old stuff. Isn't the english language over 1000 years old? What about Art? Or classical music? Or history - some of history is very old.

Well, Rhett, if you wanted a class that would be 1000-year-old English you'd be learning the language of Beowulf and Caedmon, though that was definitely a thing in English degrees of yore (see what I did there?). The problem is not that we cover old material in physics (Archimedes' buoyancy principle is how old?) but that nothing more recent is covered.

With that said--yeah, how are you really going to discuss anything modern without a smattering of basic physics terminology? Otherwise you just get a mess of concepts with nothing to string them together. And believe you me, much that I enjoyed learning Quantum and Relativity in my undergrad career, I barely had time for all my classes as it was. I can see exactly why all the engineering advisors at Union recommend dropping the "look at the pretty physics" course.

Of course, some work can be done in a traditional physics course to tie to things we know and love. Some ideas:
1. Have "Physics for Future Presidents" or "A Short History of Nearly Everything" as a supplemental reading assignment for the physics class--those books do well to tie concepts to what's being done, and is a nice assignment break from "calculate the resonant frequency..."
2. Chaos and solitons! My research has a lot to do with the latter, and honestly everything I do can be easily tied to some basic freshman physics. Fractal basin boundaries are a nice bit of "gee whiz pretty" physics that you can give some insight into with discussions of the potential energy landscape.

Though, of you want to be pedantic, solitons have been known since the early 1800s...

I'm glad you're addressing this, and I agree with some of what you say, but I have to disagree with you overall. The standard physics undergraduate curriculum is VERY old-fashioned, and it doesn't need to be that way. Like you said, even the "modern" physics classes usually mean 1920's level physics. The core 300-level classes are basically the same stuff all over again, just with a higher level of mathematical rigor. You might take a few elective classes on actual modern physics topics, but they're really light. You basically don't learn any real modern physics until you're in grad school.

The undergraduate curriculum overall seems designed to train future theoretical physicists, with hardly anything that will prepare someone for a career in engineering or even experimental physics. For example, lasers and transistors are barely mentioned except in a VERY superficial way. Even programming is barely taught at all. At my school, we had one optional programming class and that was it.