Is it plausible to do theoretical research while working?

[osnarf]

I really like physics, and I really like engineering. I like theory and I like building things and understanding how things work and all that jazz. So it's always been a tossup, hence I'm working on 2 BS degrees in Physics and EE ( the reason is because i couldn't imagine not doing either one).

I definitely plan on pursuing graduate education. What it will be in, I am still debating. While it would really be great to get a PhD in physics, get a job as a professor, and do theoretical research, I'm aware that realistically, this is unlikely, unless it turns out I'm a genius. Also, the post doc period would be rough, as I've been in a relationship for 3 years and by that time in my life I hope to be starting a family. My gf is about to start a DPT program, so if we had to move every 2 years it would be hard on her, as well, because (depending on the distance of the moves) she would have to keep getting re-licensed and tested in different states. So, unfortunately, I feel that path is out. Like I said though, while I am partial to physics, I also am very interested engineering. I am wondering if I can have the best of both worlds:

1-Finish both ugrad majors.
2-Immediately go into Ms/PhD program for EE.
3-Get a job in industry.
4-Either self-study (like I am doing with math now), or pursue masters (thesis option) program in Physics. I will already have the bachelor's degree in physics, I'd just have to continue my education.
5-Treat theoretical physics as a 'nighs and weekends job,' or a hobby.

As long as I kept up with the most recent findings, then I don't see why I wouldn't be able to have the potential to do some independent research in my spare time. I'm not talking about nobel-prize winning research, but it seems plausible I could still learn new things and contribute to physics. It would just take longer to get results because I wouldn't be able to dedicate the same amount of time as a full-time researcher would.

------Questions----
-In (4) above, would self-study of all the same topics covered in a MS be sufficient in order to have a good enough understanding to do what I'm talking about. If not, would a MS with a thesis option be sufficient, or does it really require a PhD? As I understand it, a PhD is basically a certificate that you have proven you know how to do research, but do you really need to get one to be able to do (good) research?

-What are the difficulties of pursuing a masters/PhD in something (physics) not related to one's current career (EE)? I imagine you'd have a hard time convincing your employer they should pay for it, but is it plausible to take the time off to pursue it and still have a job when you finish, provided you keep updated in recent findings in your current field?

-Do you have any other advice/thoughts on the matter?
---------------------

Thanks again everyone.

 

[AlephZero]

It depends what you call "research", but there are certainly opportunities for doing research-like work in engineering in industry. You could expect to have to earn some credibility doing project-type work first, though.

For example the (multi-national, based in the UK) company I work funds research programmes at several top-tier UK universities (including both Oxford and Cambridge) and of course want to manage what are getting for our money and keep track of what the postgrads are doing (and stop them going off down interesting but irrelevant side alleys, sometimes). We also participate in, and sometimes lead, EU-sponsored "applied research" projects involving several European companies and universities across Europe. (A typical project size might would be 30 or 40 people at 15 or 20 different companies and universities for 3 or 4 years)

Often the results of this type of work are commercially sensitive, and therefore not published fully in the open literature, but it's still "research".

 

[petergreat]

A PhD is the minimum requirement for research in theoretical physics. Theoretical physics is a hugely involved subject. It's not for amateurs. It won't work out. I agree with the other poster that industrial research is the way to go in your situation.

 

[osnarf]

AlephZero:
Interersting post, but I think we're on different pages, that wasn't exactly what I was talking about. I was talking about being employed as an engineer and doing physics 'research' that doesn't require a lab in one's off-the-clock hours.

-----------------------------------------

petergreat:
What is the difference between a masters and a PhD, except for the research? The master's degree let's the student take all the same classes (except, at my current school, QFT II). So then if the master's student fills in that lst gap, he or she has the same knowledge base to expand upon as the PhD student does; the PhD student just goes further by actually doing the research. So what's to stop a someone with a master's to do the same, just outside of the school/work?

I understand more why a PhD is required for experimentalists to learn how to and actually do their research and carry out experiments, that sort of stuff would be really hard to learn/do on your own on your own without advanced and expensive equipment. I also understand why it would be impossible to get a paid position conducting one's own research without a PhD. I don't understand why someone with all the proper knowledge base couldn't read papers published by experimentalists in a certain field and use those papers to try to put the peices together, if you will. Or even just try to solve existing problems that have been posed by other physicists.

Maybe you could elaborate?

-------------------------------

Thanks again.

 

[petergreat]

petergreat:
What is the difference between a masters and a PhD, except for the research? The master's degree let's the student take all the same classes (except, at my current school, QFT II). So then if the master's student fills in that lst gap, he or she has the same knowledge base to expand upon as the PhD student does; the PhD student just goes further by actually doing the research. So what's to stop a someone with a master's to do the same, just outside of the school/work?
Thanks again.

I understand what you're saying. As a PhD student myself I may not even qualify to give an opinion, but several things come into mind:

1) Even though a Master's degree allows you to take advanced graduate courses giving you good general understanding of theoretical physics, a PhD project is essential for you to become an expert in a specialized topic of current research. The guidance of mentors is very important, because beyond a certain stage, no textbooks exist. You need your advisor to tell you which papers / reviews to look at, and answer your questions should you have problems grasping them, because these papers are written for an advanced audience, and assumes knowledge of the subject which in turn cannot be found in textbooks. (I won't be surprised if the QFT II course you mentioned above contains essentially nothing that's not more than 30 years old. In fact that's the case for most textbooks on QFT.)

2) Research in theoretical physics is a very social activity. You need to have connections in the field. You need to exchange ideas to make progress. You need to go to seminars to learn about recent advances. Without interaction with the community you can easily go astray and become a crackpot, or get into an idea without realizing it was actually tried 20 years ago and disproved. Without institution affiliation, your ideas and talents may fail to be properly recognized, turning you into a bitter person.

3) It's been a long time since I hear about any hobbyist making valuable contributions to theoretical physics. (Garret Lisi? I'm afraid he's not taken seriously by the mainstream.) I often hear stories of people who became sick of academia and pursued research at home instead, but most of them stopped being scientifically productive. I also hear that in community colleges, where teaching leaves you little time for research, there are many delusional people who still see themselves as researchers, but never produce anything important. (Do I need to mention that these academia dropouts and community college teachers at least have PhDs.) Come on, theoretical physics is very demanding and time-consuming. When you're getting a family and a full-time job, it doesn't make sense to spend every last minute of free time on physics.

In short, I don't think working on theoretical physics as a hobby (and aspiring to make real contributions) is even an option. But it's certainly possible for you to keep an interest on physics and read about new findings even on a technical level. And a job outside academia have many benefits (I agree with you that postdoc period is terrible), like enjoying more time with your family.

 

[twofish-quant]

As long as I kept up with the most recent findings, then I don't see why I wouldn't be able to have the potential to do some independent research in my spare time.

I do, since I've tried it. Now none of these problems that I'm going to mention are *fundamental*. There is nothing that I've found that says it can't be done, but the pieces aren't together yet. They may be in a few years, and what I've been doing is to try to put together the pieces so that I can do independent research.

What spare time?

You aren't going to be at the office all the time, but when you get home, you are just going to be too tired. Also academia really isn't set up to deal with independent research. The problem is that the basic unit of research is the peer reviewed paper, and getting a paper together takes about three months of sustained effort.

The other problem is that in order to do research, you need to be highly social, and right now the social networks are just barely there. This isn't to say that they *can't* be there. It's to say that they aren't there.

One of the ideas that I've been toying with is the "National Science Reserve." If Obama gives a speech saying that it is scientifically vital that Ph.D.'s work on some projects, then I can take that speech and give it to my boss and get some time off.

I'm not talking about nobel-prize winning research

Well, Einstein did it. One thing that helped him a lot was that being a patent clerk was apparently a 9 to 5 job. The trouble with today's technical jobs is that there are incentives for the employer to squeeze as much work out of you as possible.

The other thing that Einstein did was to be very, very involved in discussing things.

it seems plausible I could still learn new things and contribute to physics. It would just take longer to get results because I wouldn't be able to dedicate the same amount of time as a full-time researcher would.

You run into the threadmill problem. By the time you get your results, someone else has already gotten them and the field has moved on.

(4) above, would self-study of all the same topics covered in a MS be sufficient in order to have a good enough understanding to do what I'm talking about.

No. This comes up to one of the misconceptions of the Ph.D. The Ph.D. doesn't teach you any facts. The Ph.D. teaches you how to do research, which is like riding a bicycle. You can read as many books on riding a bicycle as you want, but you aren't going to be able to do it unless you just ride the bicycle.

Ph.D.'s are weird because for the first time, you are finding stuff out that isn't in any textbook. That's also useful when you are trying to do stuff that no one has ever done before. I'm trying to become an "independent researcher" and I'm finding a thousand road blocks in my way. So I just look at each road block and try to figure out a way around it. You get that sort of experience when you write your dissertation. You don't get it in any master's program.

As I understand it, a PhD is basically a certificate that you have proven you know how to do research, but do you really need to get one to be able to do (good) research?

In theory no. But I don't see the point in avoiding it. Universities need a ton of serfs to grade papers, so it's a good experience. Also, you'll need a ton of social networks to do research, and being a graduate student is a great way of meeting people.

There's also the credibility aspect. If I didn't have a Ph.D., and I screamed "ACADEMIA STINKS" people could very easily ignore what I say. The fact that I have that piece of paper puts me inside the club so to speak, so that when I say "ACADEMIA STINKS" people can't say that I'm a nobody. One other cool thing is that if you have someone with street cred saying something, that gives people without credentials "permission" to say the same thing.

What are the difficulties of pursuing a masters/PhD in something (physics) not related to one's current career (EE)? I imagine you'd have a hard time convincing your employer they should pay for it, but is it plausible to take the time off to pursue it and still have a job when you finish, provided you keep updated in recent findings in your current field?

It depends on the Ph.D. There are some fields where it is standard practice for working people to get Ph.D.'s, but physics isn't one of them.

Also, the way that companies are set up, it's really hard to get time off, because it's cheaper for the employer to replace you with someone that doesn't want/need time off. One reason I think it's important for there to be some structural changes is that if you had a "National Science Reserve" you'd also have laws that forbid employers from using time off against you in the same way that employers can't fire you if you are on jury duty or doing Army Reserve training.

The other thing you could try is to find a field in which you make a ton of money, keep eating ramen for dinner and retire early.

 

[twofish-quant]

Even though a Master's degree allows you to take advanced graduate courses giving you good general understanding of theoretical physics, a PhD project is essential for you to become an expert in a specialized topic of current research.

Also it teaches you how to quickly become an expert if you aren't one already. One reason investment banks hire physics Ph.D.'s is that it's assumed that they give you a general problem on a topic you know nothing about, that you will very quickly do whatever you need to in order to be an expert in the topic.

The guidance of mentors is very important, because beyond a certain stage, no textbooks exist. You need your advisor to tell you which papers / reviews to look at, and answer your questions should you have problems grasping them, because these papers are written for an advanced audience, and assumes knowledge of the subject which in turn cannot be found in textbooks.

And then there is the emotional support. Some times when you feel rotten, you just need someone to tell you that you are doing OK. You've just made a fool of yourself in a presentation, it's obvious that your approach doesn't work, everything is burning, you feel tired and depressed and you just want to give up.

Having someone tell you that this is "normal" is really useful.

The other thing is that at some point, your mentor will let you know that they can't offer you much useful advice because you know more about the topic than they do. Also you pick up culture and assumptions. If you are around a person of authority for long enough, you unconsciously start thinking and acting like they do.

Also this lasts a lot longer than the Ph.D. Graduate education can be compared to being thrown into the ocean and forced to swim to shore. The good thing is that you know that there is someone there with a life preserver that can pull you out if you start to drown, and once you make it to shore, then you can celebrate that you made it there.

But once you've been dunked in the water, it trains you for situations in which your teachers aren't there any more. Now there are no life preservers, but you can think to yourself, I made it once, and if there is any way of making this work, I'm going to find it.

(I won't be surprised if the QFT II course you mentioned above contains essentially nothing that's not more than 30 years old. In fact that's the case for most textbooks on QFT.)

And in some fields it's worse than that. In some areas of finance and computational astrophysics, anything that is older than six months old is hopelessly out of date.

3) It's been a long time since I hear about any hobbyist making valuable contributions to theoretical physics.

Albert Einstein. Got his Ph.D. Worked as a patent clerk.

But that's the cool thing about getting a Ph.D. Once you've done one thing that no one else in the world has ever done, that doesn't stop you from trying to do something else that no one else in the world has ever done.

And sometimes the world changes. If the problem is social networks, well facebook didn't exist ten years ago.

I often hear stories of people who became sick of academia and pursued research at home instead, but most of them stopped being scientifically productive. I also hear that in community colleges, where teaching leaves you little time for research, there are many delusional people who still see themselves as researchers, but never produce anything important.

Yup. But that's where the "Ph.D. mentality" is useful. Someone without research experience will look at the fact that everyone before them has failed and give up. Someone with research experience will look at all of the cases in which something didn't work, think about it, try something different which may or may not work, and then keep trying until something useful comes out.

If the problem is time and social networks, then what can I do to get time, and how do I get social networks.

Come on, theoretical physics is very demanding and time-consuming. When you're getting a family and a full-time job, it doesn't make sense to spend every last minute of free time on physics.

So what can we do to save time?
Why can't you survive on a part time job? (Yes, there are reasons, but it's worth asking the question).

In short, I don't think working on theoretical physics as a hobby (and aspiring to make real contributions) is even an option.

What can be done to make it an option? Let's go down each of the reasons why it won't work, and then stare at it to see if there is a way around the road block. It might take a few years, but progress is always difficult.

Also the world changes. What isn't possible in 1950 becomes possible in 2000, and one of the cool things about research is that sometimes what is impossible in January 2010 suddenly becomes possible in June 2010.

The basic problem is time, money, and social networks. None of these things are insurmountable. If you can give me an argument that being a physics hobbyist violates Lorenz covariance, then that's fine, but no one has. The barrier that people have mentioned are real, but they aren't anything that can't be changed.

One thing that you have to realize about being a physicist is that in some ways you can't stop being a physicist. It's not like being a factory worker where you can clock out. It's more like being a priest or a cop, in which you are always a priest or a cop. Once you start looking at the world in a certain way, you really can't stop.

 

[CronoSpark]

There's a huge gap between bachelors and PhD. I found that once I got into graduate level research, I was literally in a swamp or if I wanted to catch up to the "latest and greatest" it was tough to do so even spending plenty of time on the topic. Like the previous posts, there is indeed a huge gap/void that you will feel and to speed things up you will need helping hands (social networks/supervisors/etc...). Research happen fast too, so there's no time for breaks.

However, if you really insist of trying physics research on your spare time, I'd suggest really narrowing on a very specific topic of interest. Simplify it as well. Contributions don't have to be revolutionary, but you will go through the process of what it takes. Not having a PhD or Masters in the field would probably decrease your credibility by a lot though. Unless you were a well known genius to those who have influence, I'd suggest you start creating lots of social contacts with people who are in the field. You will need to gain influence through other means, basically.

Ask some of your old profs who's in the field you like, and ask them if there's any "simpler" research to do, could very well be experimental, but you will learn as you go along.

Edit:

Also, on another note, raising a family will be very time consuming. Unless your future wife is very supportive or can handle most of the child raising business herself, you will find yourself swamped until at least your child is old enough to be self sufficient. Mental stress will be a major factor for you when you are pursuing what you are proposing here.

 

[twofish-quant]

Like the previous posts, there is indeed a huge gap/void that you will feel and to speed things up you will need helping hands (social networks/supervisors/etc...). Research happen fast too, so there's no time for breaks.

Also, you learn about some of the "little things" that universities provide once you don't have them. One thing that you can easily take for granted is a well stocked research library. Now there isn't any *technical* reason, why it can't be digitalized, but there are some ***monster*** economic and political issues (basically who gets the money), and it hasn't been totally done yet. There's also the nice (or sometimes not so nice) departmental secretary that makes sure that all of the forms are filled out. If you log into to your account, everything magically works.

And there is the world of grant proposal writing and government lobbying.

You could put this online, and there are some efforts at doing that

http://www.mica-vw.org/wiki/index.php/MICA_Seminars

Again, none of this is undoable. It's just that it hasn't been done. If it's possible to topple a major government by facebook, then it should be possible to build a university with it, and it's just a matter of rolling up your sleeves and doing it.

Not having a PhD or Masters in the field would probably decrease your credibility by a lot though.

I think it's pretty essential that you get your Ph.D. It will give you an idea of what it is that you are trying to recreate. Also there's not a lack of Ph.D.'s outside academia that are willing to help you.

Unless you were a well known genius to those who have influence, I'd suggest you start creating lots of social contacts with people who are in the field.

Also the big mistake that I made was that after I got my Ph.D, I was so annoyed that I let my social network grow cold. That was a problem because it's a 100x times harder to recreate a social network than to maintain an existing one. One other thing that is going to be essential is for you to attend conferences. Face to face interaction is really important, and if you can't get it through day to day contact, you need to attend conferences.

You will need to gain influence through other means, basically.

In fact, you'll gain influence through the same way that people within the system get influence. If people think highly of your research, they'll write back with your e-mail. The way that the cycle works is "publications" -> "credibility" -> "resources" -> "publications."

The problem with this cycle is that it spirals upward or downward. There are lots of smart Ph.D.'s working in community colleges, but they get caught in the downward spiral.

One other thing is that there is a community of amateur astronomers which could be used as a model for this sort of thing. Also one reason I'm interested in the professional sports and acting industries is that the social dynamics of acting works a lot like astrophysics. It's very much a "star" system.

 

[twofish-quant]

Just to give you an idea of what it takes to "restart the engine." If I were to quit my job right now, and spend full time trying to do supernova research, I figure that it would take about three months for me to relearn all of the stuff that has happened since the network got cold. It would take probably another year for me to come up with something original and get myself to the point where I could publish something that I'd find acceptable. That's assuming that I had social networks in place, and getting those would be some effort.

The good news is that if I put myself in a position where I could to things full time, I could get myself back into shape within a year or two. One thing about my current job is that I'm doing enough hard mathematics so that I'm not getting out of shape in that area.

Also, the reason that getting a Ph.D. is important is that it proves that you've internalized some standards. I've "absorbed" the standards for what quality research looks like, so I can tell whether what I'm doing is high quality or low quality.

 

[petergreat]

Einstein's success in physics while working as a patent clerk is not only due to his extraordinary intelligence, but also due to the fact that he lived in an extraordinary time in history. Special Relativity was lurking in Lorentz and Poincare's work and was awaiting Einstein to finally see the truth. Quantum mechanics was slowly arriving, and relied upon many people's pioneering work including Einstein's theory of the photo-electric effect. The atomic theory of matter was a little short of convincing everyone, and needed Einstein's analysis of Brownian motion to prove it beyond doubt.

It's notable that Einstein's 1905 discoveries are revolutionary but also simple. Today's physics seems to allow not much room for a hidden genius to make important discoveries based on a simple elegant explanation of a complex phenomenon (photo-electric effect and Brownian motion) or a single deep insight into a problem (principle of relativity).

 

[twofish-quant]

Einstein's success in physics while working as a patent clerk is not only due to his extraordinary intelligence, but also due to the fact that he lived in an extraordinary time in history.

And we *don't* live in an extraordinary time in history?

The reason I'm giddy about computational astrophysics is that calculations that required an entire supercomputer center in 1985 can now be done on your cell phone. The fact that we have dramatically huge amounts of compute power means that things that couldn't have been done before can be done now.

Today's physics seems to allow not much room for a hidden genius to make important discoveries based on a simple elegant explanation of a complex phenomenon (photo-electric effect and Brownian motion) or a single deep insight into a problem (principle of relativity).

It does, you just need to look in the right place.

For example, in computational astrophysics, you'd like for the computer to simulate everything, but you that doesn't work. So you spend most of your time coming up with clever ways of modelling parts of the problem in simple elegant ways, so that you can find the "essence" of the problem. Or look at Lambda-CDM, explain the entire universe in twelve numbers.

There are a lot of unsolved problems in astrophysics that are amenable to insight. Magnetic fields, combustion, fluid dynamics, equations of state, etc. etc. etc. Also I'm pretty sure that there is a blindingly simple model for galaxy formation and/or core collapse supernova, that we haven't figured out yet.

Just to give people a problem which I think that there is a simple elegant solution that no one has figured out. I have a 20 year old CPU-based code that I'd like to get running on a GPU quick. I don't want to rewrite the whole code for GPU. If someone can show some sort of simple mathematical transform, or some need trick to get a CPU code very quickly running on a GPU, that would be cool.

 

[Ryker]

It's notable that Einstein's 1905 discoveries are revolutionary but also simple. Today's physics seems to allow not much room for a hidden genius to make important discoveries based on a simple elegant explanation of a complex phenomenon (photo-electric effect and Brownian motion) or a single deep insight into a problem (principle of relativity).

You don't know that, that's the mentality people have prior to coming up with any "simple" and important discovery that turns the world upside down. And everything seems simple after the fact.