Do people care about single-author works in quantum information theory?

Question

I am a junior postdoc (completed one and a half years) working in the field of quantum information theory. I was told by my PhD advisor that single-author papers don’t really matter in physics. They said that it’s all collaborations now, citing the example of experimental sciences. I was surprised by how callous this sounded.

This comment was made in relation to a single-author work of mine which is something that came out of my PhD. I pursued it myself and as I didn't really get any relevant help, I ended up being the sole author.

So I am just wondering what the general consensus is on single-author works in the quantum information community, i.e. if you are a postdoc advisor would you encourage your postdocs to write single-author papers? As a PhD advisor what would you think of your ex-students when they write their own papers?

Answer 1

I do not work in quantum information theory, but I think my insights can be generalised with some care. For whatever it’s worth, I have published two single-author papers arising from my PhD in computational physics.

First note that in most subfields of physics I am aware of, single-author papers are rare. This is simply due to the nature of academia where most work is done by a student or postdoc and supervised by a professor. This doesn’t mean however that single-author papers are not valued. Showing tons of examples of multi-author works (as your PhD advisor apparently did) doesn’t prove anything. That doubly applies if those papers were all experimental, as it’s far easier to produce single-author papers in theory.

Since your PhD advisor’s statement was about physics in general, it can also be falsified by just looking at how well single-author papers tend to do. It took me a few minutes to find some recently published single-author papers in Physical Review Letters, and I could not even filter for theoretical papers, let alone single-author papers. If you want to be sure about your particular subfield, look for single-author papers in it and check whether those that you find are regularly cited. Maybe you already cite some yourself. Just beware that since those papers are inherently rare, you really have to search till you have a statistically relevant population.

If you are a postdoc advisor would you encourage your postdocs to write single-author papers? As a PhD advisor what would you think of your ex-students when they write their own papers?

I think that this may be a false proxy for what you actually want to know. Papers do not (or at least should not) become single-author papers because you design your research like that from the beginning. Rather, you set out on a project you conceived and it turns out that you complete it without any external input qualifying anybody else for authorship. Mind that this includes the writing process for which previous publication experience is extremely helpful, if not essential.

That being said, suppose I supervise somebody and the above happened: They conceived a project and completed it without any relevant input by me or anybody else (but with me being aware of it, assuming I am their boss). If I do not see any flaws (which of course would be relevant input), and they have the skills to handle the writing and publication process on their own, I would advise them to do exactly that. I would ensure that they have access to constructive criticism along the process, e.g., I would offer to read the manuscript (internal peer review), unless they have somebody else for that, hoping for nothing but an acknowledgement. One reason to encourage this is that they can demonstrate that they can perform and publish research on their own (which is what a PhD should qualify for), which can be crucial for their career.

Of course, you should be wary that this does not translate to all advisors and situations. Some people think they need to be the author of everything published by somebody in their group (violating authorship ethics). Some people might also get legitimately angry that you pursued a project without telling them anything about it, displaying a lack of trust or transparency towards their boss (in the work sense).

---

Further reading on this:

• Does one really need a co-author to avoid being considered a nerd?
• PhD students publish without supervisors – how does it work?

Answer 2

In addition to the other answers that cover the value of publications I would like to point a possible misunderstanding here.

"Single-author papers don't really matter in physics" can be understood as a contrast to "Single-author papers are very important in field X". I have seen reports of applicants for professor-positions where the lack of single-author papers was pointed out as a shortcoming of one of the applicants. This has nothing to do with the quality of the papers. In some disciplines being able to write a single-author paper in itself is seen as a strong point. And not having any can be thought of as lacking independence. Your advisor may just have wanted to let you know that this is not the case in physics.

This has consequences for what a good strategy to getting tenure is. If time allows you to write 2 papers with 3 authors or 1 single-authored paper, the best (to get your next position, not necessarily to do high-quality research) choice depends on the field. And it seems that in your field collaboration is more valued than single authorship.

Of course I don't know if that was what your advisor meant, but it may be worth asking them.

Answer 3

I am quite sure that across all scientific disciplines, it is first and foremost the quality of the work that is relevant, and not the number of authors. So if you are able to find a relevant research question and come up with a solution all by yourself, there should be no problem whatsoever to write a single authored paper about it. What would be the alternative? Give somebody else a gift authorship to make it appear a collaborative effort? That would certainly not be the right way to go.

If a single authored paper does not get much attention, it will be more due to the contents of the paper which does not matter to others much. In fact, in certain disciplines it might be difficult to find open and high impact research questions that can be covered by a single person, so maybe that is more what you supervisor was getting at? This may be because the field is old/mature, or it is so super complex that one would expect to require at least a few highly specialized individuals to make progress, among other reasons. In this line of thinking, a single authored paper could then be perceived a priori, i.e. without reading it, as a paper with a higher probability to address a relatively simple research question with a relatively simple methodology.

Something else to keep in mind: If your publication list consists of mainly single authored papers, it will make the impression that you prefer to work alone or are unable to collaborate with others. However, generally speaking, collaborations make a lot of sense in research, and your (documented) ability to work with others will be relevant for future employers.

Answer 4

This is not specific to quantim information theory, but I see the following dangers associated with single author work:

• Your work is easily ignored by others

  They might not get aware of your work, or do not fully understand the impact and value of it. Especially if it is a new or slightly different approach. This may easily happens if you work alone. Also in my experience only junior researchers actually perform a literature research. (Established) senior researchers mostly get aware of adjacent research by personal contacts. Having a (well known and established) senior researcher on your publication gets it attention of his/her peers. Maybe it should not be the case, but practically that is what happens. There may be examples where this is not the case but an isolated researcher has certainly no advantage over a well integrated, collaborating researcher of the same skill.

• Your research may only be tangential to the (current mainstream) research (and you cannot really influence the mainstream direction)

  Even though your work might have impact, convincing others of the importance may be hard. It will not happen automatically. Collaborating with others will ensure that you use the same language/mindset of others and make it easier for a community to connect your research to the current mainstream. Also you can influence the community much easier when you involve more people as if you are alone.

Apart from that I see no immediate disadvantage from single author publications. Especially when it comes to theoretical research I have a hard time seeing a large number of authors on a publication like for experimental works. How should that actually work?

But being an "isolated" researcher, for which single author publications is a symptom, is a risk I would say.