Why is Griffiths not a good introduction to QM?

[nonequilibrium]

I've seen quite a few people on this forum agree that Griffiths is not a good book to learn QM from.

But I was wondering what reasons people would give for it.

(Personally I don't feel I can give an in depth critique since I've only sampled his book. I rather liked his explanation of perturbation theory, however he lost my interest when he made a--in my opinion--fatal error in the first chapter when he stated that (all) realistic interpretations have been proven wrong. However, I don't think many people would care about such matters, so I'm wondering what other peeves people have about the book.)

 

[Nabeshin]

I can see two reasons people would object:

1) The book starts with the Schrodinger Eq. For someone who has never seen this equation, its motivation, or the historical development of QM (via old QM), this is a bit puzzling. It's not something that's 'intuitive' like F=ma, so I can imagine seeing this for the first time is not the greatest way to build intuition for QM.

2) The formalism is skimpy. As an upper-division QM book for many schools, it feels a little light on the linear algebra and deeper mathematics, much more content to simply *do* things. But I guess this isn't really a knock against an introduction text, it just puts the book somewhere in an awkward place between an intro wave mechanics text and a book which goes much deeper into the formalism.

 

[Jorriss]

1) The book starts with the Schrodinger Eq. For someone who has never seen this equation, its motivation, or the historical development of QM (via old QM), this is a bit puzzling. It's not something that's 'intuitive' like F=ma, so I can imagine seeing this for the first time is not the greatest way to build intuition for QM.

2) The formalism is skimpy. As an upper-division QM book for many schools, it feels a little light on the linear algebra and deeper mathematics, much more content to simply *do* things. But I guess this isn't really a knock against an introduction text, it just puts the book somewhere in an awkward place between an intro wave mechanics text and a book which goes much deeper into the formalism.

These are both fair complaints. Griffiths spends one chapter on formalism then more or less drops it. I think he should, in that chapter, also presented QM from a postulate pov.

I also think he is too chatty. I think his E&M book has a better tone. It's not super formal like a math text but it's not as nonchalant as his QM text.

 

[chill_factor]

I don't like it because there's not enough information to answer the problems in the book.

I have no idea how he got some of the answers he got, using information only presented in the book, especially in chapters 4 and beyond. There is also no complete solutions manual with explanations.

There are also very few examples.

However, seeing as the other choices at the UG level are probably much worse (only good one I've seen is grad level Zetilli, which has hundreds of example problems) I believe that it is the best choice.

 

[bhobba]

I agree its one of the best of a bad bunch. Its problem is it does not teach it in a systematic way from the math to the postulates to an interpretation to solving problems. By far the best book IMHO is Ballentine but it is at a graduate level:
https://www.amazon.com/dp/9810241054/?tag=pfamazon01-20

I think Griffiths is OK as a warm up to Ballentine but I would supplement it with the Structure and Interpretation Of QM by Hughes:
https://www.amazon.com/dp/0674843924/?tag=pfamazon01-20

However IMHO Griffiths is overpriced and I would use David McMahon's book instead:
https://www.amazon.com/dp/0071455469/?tag=pfamazon01-20

Thanks
Bill

 

[vanhees71]

Ballentine is also one of my favorites. As an undergrad text, I'd recommend to look at the Feynman Lectures vol. 3. As a more advanced book, which we have used in our first course on quantum theory in Germany, and which I still like today is

J.J. Sakurai, Modern Quantum Mechanics.

A good book concerning a lot of interesting applications is Landau/Lifgarbages vol. 3.

A kind of bible in the path-integral approach is

H. Kleinert, Path Integrals in Quantum Mechanics, Stastics, Polymer Physics, and Functional Methods

 

[genericusrnme]

1. The difficulty of the problems varies from quoting earlier text to problems the textbook does NOT prepare you for.
2. Inconsistencies in definitions
3. Little to no motivation (extremely bad when you look at the ladder operator for the harmonic oscillator section, god forbid you learn about symmetries)
4. The phrase 'mathematicians tell us the answer is'
5. No path integrals

And that's just scrathing the surface..
It's a shame because his intro to electrodynamics textbook is pretty good.

I'd suggest Shankar's Principles of Quantum Mechanics for a more modern treatment or Landau and Lifgarbagez vol. 3 for a slightly older 'wave mechanics' based treatment. Both have just the right balance of exposition and rigor for a subject like quantum mechanics.

I'm not sure why people reccomend Ballentine, I found Shankars book much nicer but each to their own I guess.

 

[mundumustafa]

Hi,

Most problems in Griffiths can't be solved with ideas Grifiths gives in his book (i usually use Mahons ideas to solve them)
Not only Shankar's book but even his video lectures on QM are superb.

MM

 

[ahsanxr]

I'll quote another user from an earlier thread that got locked. I agree with pretty much everything he has to say.

Hello everyone.

I wanted to make this thread to asks lecturers to please stop using Grffith's book "Introduction to Quantum Mechanics" as the required text for undergraduate physics, and to warn students about it. Please read this, as I truly believe Griffiths' book is a rip-off, nevermind a bad book, and that there are far superior introductory QM books out there. Let me explain.

Firstly, there is the price. Over $100.00 is ridiculuous when there are many good introductory QM books that are roughly half the price: Shankar, Baym, Rojansky, Bransden & Joachain, Zettili,.. you get the idea. I will return to this topic later.

Secondly, you don't get much book for your buck. Griffiths seems to want to explain everything with the least amount of words as possible. He states that the book does not cover enough material to make a full year course. I find this difficult to fathom considering the price and the size; surely he could have added more topics?

Thirdly, most importantly, this book is a horrible pedagogical text. For this book one merely needs to know how to take partial derivatives. In other words, anyone who has passed their first year could read this book. You may think this is fantastic, as it makes the book simpler to read, alas, it does not. For instance, Griffiths explanation of the separation of variables technique is painful. The only reason I could follow his writing is because I had done partial differential equations prior to the course. Those who had not were confused by this. Griffiths does not even expect you to know linear algebra (it's all in the appendix) or to solve an ordinary differential equation-- he always supplies the answer and tells you to check that it is correct by substituting it back in the formula.

Griffiths does not explain anything. At the very beginning of the book the Schrodinger equation is thrown at the student. Not much is said about it, what it means, the justification for it, how this departs from classical mechanics-- Griffiths does not say. As a result one does not know what they are doing or why. Isn't the reason we do physics is because we enjoy it? Griffiths does not motivate anything. He says ''Here's a situation that occurs in QM, here are the equations relevant to this situation, and mathematicians tell us that this is the answer'''. Read the book, it really is like this. This is very much a rubbish way to teach anything. Griffiths never even properly explains what the wave function is all about. When I look at this book this is what I see: here are a bunch of differential equations that occur in QM, and if you told a mathematician to solve them this is what you get as the solution. The book is horrendous.

Griffiths exposition is terrible. Many pretty pictures are fine, but I want substance. He leaves out much theory that would make QM easier and more understandable. His organization is terrible. He introduces very little theory at first, and gradually introduces the theory throughout a few badly explained examples. What you get is an incoherent theory scattered throughout he chapter, and no overview of what you have done. Other books present general theory (and more theory) before the examples, which serve to illustrate the theory. After the examples more theory is given, and this is how it should be.

QM is simpler with more maths, as Griffiths himself admits, but the only prerequisites for his book is taking partial derivatives. QM is much, much simpler if the author assumes the reader has knowledge of linear algebra, advanced calc, ODEs and some elementary PDEs. Students who do not have this math background SHOULD NOT BE STUDYING QM!

Griffiths provides very few examples in each chapter, and leaves essential ideas and theorems to examples. There are no answers/solutions to the problems, and I have no qualm with this. Except, if you are going to require very fundamental ideas are left as problems, the student can get a totally wrong set of fundamental principles if left unchecked by not having the answer. Also, many of Griffith's questions are very vague, and one struggles to figure out what he is asking. Many of the questions are stupid mathematical arguments.

Griffiths is inonsistent. He does not expect you to know integration by parts (he explains it in a footnote), but expects you to understand terms like "canonical substitution". At one point he defines the transmission coefficient as |F|^2/|A|^2, and then in a later problem he says "prove that in this instance the transmission cofficient is not simply |F|^2/|A|^2. This occurs in chapter two. He does define it this way, he uses the equality sign with three lines. This sort of inconsistency occurs throughout the text. If Griffiths had simply defined the transmission coefficient correctly at the start this would have been avoided. His theory is patchy and he does not give all the theory you need to solve the end of chapter problems, or even to understand QM correctly.

I have more problems with the book, but at present they elude me. There are so many faults and short-commings I cannot recall them all at present. When I read the book again I will surely spot more.

Please, please, I beg you, do not use Griffiths. Griffiths is for people who want to look cool posing with a book on QM. Two excellent books: Zettili and Bransden & Joachain are both far superior and $40 cheaper. I understand profs may think Griffiths is simpler since it requires less maths, but as I explained, this is part of what makes it a bad book. All the other good undergrad QM books use much more maths and are much better for understanding and technique; all-round much better books. Profs also like Griffiths since it has no solutions. Try, for one year, to use Zettili or Bransden & Joachain instead of Griffiths, or get yourself to use Griffiths and one of the other books, and judge for yourself. Bransden & Joachain cover much more material. Zettili has worked problems. Both give more theory, explain everything much better, and just have much more content. Zettili is at 650 pages and B & J over 700 (in the first edition). Griffiths books has roughly half as many words per page too; I know, I've counted. Even the paper Griffiths book is printed on is inferior. Please stop using Griffiths QM, this is a passionate plea from an undergrad. I truly believe Griffiths wrote this book to make money. There is very little content and substance to this book, especially for the price and compared to other undergrad QM books. A first year student could read the book there is so little mathematics. I believe he did this on purpose to entice ill-prepared over-zealous young students. Please stop using it.

 

[lugita15]

As an undergrad text, I'd recommend to look at the Feynman Lectures vol. 3. As a more advanced book, which we have used in our first course on quantum theory in Germany, and which I still like today is

J.J. Sakurai, Modern Quantum Mechanics.

That's why I like Townsend's A Modern Approach to Quantum Mechanics. It's basically like an undergraduate version of Sakurai, so that it does a good thorough job of introducing the mathematical formalism like bra-ket, while demanding less in the way of mathematical prerequisites. But what I really like about it is that, like Feynman Lectures volume 3, it gives the physical motivation in terms of spin systems, so that you work your way through finite-dimensional Hilbert spaces before getting to time-evolution and continuous positional eigenstates.

 

[lsaldana]

That's why I like Townsend's A Modern Approach to Quantum Mechanics. It's basically like an undergraduate version of Sakurai, so that it does a good thorough job of introducing the mathematical formalism like bra-ket, while demanding less in the way of mathematical prerequisites. But what I really like about it is that, like Feynman Lectures volume 3, it gives the physical motivation in terms of spin systems, so that you work your way through finite-dimensional Hilbert spaces before getting to time-evolution and continuous positional eigenstates.

This. Townsend is a great book to learn QM for at the undergraduate level. It really is a baby Sakurai in many ways. Making the transition between these books was natural for me. The formalism presented and the mathematics demanded are fair and just right for an average physics major. It's mostly linear algebra and less so differential equations and to the trained eye, the book screams group theory throughout (for example see Ch.3).

As much as I like Griffith's E&M book, the QM book just doesn't cut it. Don't get me wrong, I really enjoy his tone throughout the book (not to mention his off-hand discussions on the footnotes). However, it mostly just shows QM in one of its facets: position space. This, I feel, is it's greatest weakness. The one chapter on formalism was interesting but not enough and is vaguely used in the remainder of the text. Once you learn that all the wavefunction stuff Griffiths has been pushing at you all year, is just a, say, "subset" of what's out there, you'll be left wanting more. Simply put, there's more to the story than just the wavefunction and you won't get much more than that in Griffiths.