1. Jul 16, 2011

### shanesworld

I wanted to gain a sense about some of the opinions floating around out there about publishing works with out the use of references. Of course, the vast majority of papers certainly include references these days. However, in principle it is still possible to write a paper that does not directly borrow materials from another source, or that is original/based on first principle, etc. For instance Einstein's initial paper on special relativity did not include references. But, today I don't see too many, if any published articles without references. So, for the sake of the pole, I'd like to see where some people may sit on the topic of papers that do not directly borrow from other works. I could see including a list of resources outside of the paper that address similar topics to things you are discussing. But if you did not directly utilize these papers in your work, should they be included as "references?" Additionally, is there an assumption that papers are "supposed" to include references? If so is this a functional assumption in every case? Does anybody have any experience about publishing when they didn't feel references were necessary for their paper?

2. Jul 16, 2011

### DragonPetter

I think references just give more credibility to the statements you make in your paper. If you don't use a reference and there exists material on it already, people may assume you really didn't put much thought or research into the topic you're talking about which takes credibility away.

I don't think there's any rule that you can't publish without references, and of course, if you come up with a new idea, that stuff will not have references.

I think its actually kind of annoying when authors list too many references, and it feels like they are using references for almost a cosmetic reason.

3. Jul 16, 2011

### Staff: Mentor

What Einstein was able to get away with 106 years ago has little bearing on current publication practices.

4. Jul 16, 2011

### Staff: Mentor

References help the person reading the paper by providing evidence for claims. If I were to write "20ng per ml of PMA has been shown to be the most effective at causing monocyte-to-macrophage differentiation due to oxidative stress" I should back that up. Firstly so that people could know it is true and secondly so they can examine why it is true. If I didn't provide a reference I would have to provide the data itself to back up the claim, this would be incredibly impractical for studies involving intricate methods that have been refined by decades of other papers. It's much easier to say "total DNA assays were performed using protocols previously described [REF]" than "the following 80 pages of data summarise the data gained over the past decade that explains the parameters of the method"

5. Jul 16, 2011

### Pengwuino

To add to what ryan said, no one is going to be interested in reading a paper 100 pages long. That's why when you're writing a paper and you need to use some piece of information but don't want to do the whole calculation that someone else did, you simply cite them and say "this is shown by *whoever*".

Maybe someone does a nice calculation on something and in the years that follow, 100 people eventually want to cite that calculation for whatever reason. Why waste the time on those 100 new papers writing out a calculation when they could all save hundreds of pages of journal pages and simply cite your paper?

Also, if I were a referee, I would be highly highly HIGHLY suspect of a paper with no references. A vast majority of crackpots who walk into PF have claims that reference 0 other current works. This means that they haven't even taken the time to find out what the current state of the science that they're doing is! Some people walk in with theories that were refuted 80 years ago! So no references implies no actual research which strongly implies it's garbage.

6. Jul 16, 2011

### Staff: Mentor

And to add to what Pengwuino has said published papers need to not only show data, outline the methods and discuss the relevance, they also have to provide the reader with a background knowledge to put the study in context. Even experts in the field need to have it clarified as to why you are doing the work. It would be hard to justify some biomed research if you don't have an introduction outlining why there is a need for the medicine, what the current incidence/prevalence is, what current treatments are plus their efficacy and costs etc etc.

7. Jul 18, 2011

### shanesworld

.
It appears that most of the replies on here are strongly favorable towards references. I quoted the above person because it pertains most directly to the questions in my post about the level of assumption out there. I, having always included references in published academic works of my own, am well aware of there value. However, I must also admit there are times when I have included a reference in a paper when I didn't truly consider that crucial to my work, simply because I felt pressure to have referenced papers that adressed related topics. Of course, better safe than sorry.

Let's agree that the background section of a paper usually discusses the current state of research, which most likely means a discussion of other authors/ researchers development, i.e refences are probably needed here (in most cases). So people who fail to do that are risking shooting themseves in the foot by assuming their work is original, when in fact they needed to do their "homework" more.

But to get back to my original post, the pole is also about, material which is not credited to other researchers directly, perhaps original, or perhaps use of things that are just common knowlege, Newton's laws, basic geometric identities, and such. I would usually think it is not necessary to say things like, "it is well known that by Newton's second law F=ma" and have to find another paper to reference for said statment, but I could be wrong.

Last edited: Jul 18, 2011
8. Jul 18, 2011

### shanesworld

Agreed. In fact that is well put...I also see the merit of including references simply so that readers have the opportunity to read more about topics/ related topics to what an author is discussing. Accessibility is always helpful.

9. Jul 18, 2011

### Staff: Mentor

I agree with ryan_m_b. I don't think it is possible to write a good introduction or discussion section without using references, even if your methods section is completely novel (which is very doubtful).

10. Jul 18, 2011

### Pengwuino

There is a line that does have to be drawn as to stuff that is considered so "well-known" that you don't cite it. The guide as to where the line needs to be drawn basically comes down to who will read your paper. If you're submitting it for publication known for gravitational research, you don't need to tell people that the Einstein equations are $G_{ab} = 8\pi T_{ab}$; they already know it. Assume your audience is grad students/phd educated researchers.

If it's something original, there's no one to cite in the first place.

To add to this, remember every good physics problem has it's problems in the details of that problem. Even vague, completely open problems such as dark energy and dark matter have details that any good theory needs to be able to work out. So no matter how groundbreaking or novel an idea is, it still has to answer the questions posed by whatever problem you're working on. To do that, you're pretty much forced to cite other work that at the least tells you what the problem really is.

To me, this is convincing enough for me to say a good research article is wholly incompatible with having no citations or references.

11. Jul 18, 2011

### Staff: Mentor

I don't see a pole.

Shanesworld, can you pm me a link to at least one of your publications? I won't share it, I'd just like to see what you're talking about.

Thanks.

12. Jul 18, 2011

### shanesworld

I find this to be a nice response.

13. Jul 18, 2011

### Staff: Mentor

Shanesworld, I can't wait to see what you will be sending me. I'm excited to read some of your published papers!

14. Jul 18, 2011

### micromass

Staff Emeritus
I'm actually quite curious too as to what kind of paper doesn't need references
I mean, you can't build a theory from scratch? You always need to have some foundations. And these foundations are exactly the references...
When I wrote my thesis, I also didn't feel the need to include much references for results I found trivial. But my advisor told me to insert some "blanket references", some kind of textbook that covers most of the theory. These are references that you must have. You're doing something on group theory and you only use trivial results? Well, insert a group theory book in your references nevertheless. It isn't an essential reference, but insert anyway.

Furthermore, be sure to reference people on the same department as you. They'll like you for it :tongue2:

15. Jul 18, 2011

### Staff: Mentor

Try this one - http://www.otds.co.uk/electrical-equipment/steel-poles/steel-pole-4.jpg [Broken]

Last edited by a moderator: May 5, 2017
16. Jul 18, 2011

### shanesworld

Ha Ha, nice picture.

Last edited by a moderator: May 5, 2017
17. Jul 18, 2011

### shanesworld

For instance, http://www.fourmilab.ch/etexts/einstein/specrel/specrel.pdf
although you may debate the relevance to today's publishing formalities.

18. Jul 18, 2011

### shanesworld

Interesting

19. Jul 18, 2011

### Staff: Mentor

No response from the OP, this is closed, there are no publications that have been furnished.

shane, if you think I'm not trustworthy, pick any mentor of your choice to send your publications to.

Last edited: Jul 18, 2011