Is it bad form to use the word “novel” when describing your own work in a paper?

Question

Ideally any research paper should contain novel thoughts, ideas, procedures, observations, methods, etc. of some form or another. That being said, is it really necessary to remind the reader about it by using the words novel, revolutionary, or never been done before as opposed to just comparing it to existing studies and noting the differences between the two?

Answer 1

Yes, I would advise against using the words novel or new for the following reasons (which overlap with what you already argued):

• It’s almost impossible to fully ascertain that something is new. Even if you just discovered some outstanding and surprising effect like superconductivity, somebody else may have discovered it a month earlier and is just about to publish it.

• At the same time, you are expected to have ascertained the novelty of what you are doing as well as reasonably possible and to document it in your introduction or literature review. To quote JeffE:

  The message you want to send is "I have read every paper on this topic, and none of them do this thing that I'm about to do."

  If you have any reason to believe that what you are doing is not novel, you should either document this (e.g., “we tried to reproduce this experiment”) or not try to publish at all. Thus, there is never any reason to claim something is novel – it is implicit that everything you write that is neither common knowledge nor equipped with a citation is novel to your best knowledge.

The following memo by the American Physical Society (one of the most prolific publishers in physics) argues similarly:

Physical Review adheres to the following policy with respect to use of terms such as "new" or "novel:" All material accepted for publication in the Physical Review is expected to contain new results in physics. Phrases such as "new," "for the first time," etc., therefore should normally be unnecessary; they are not in keeping with the journal's scientific style. Furthermore, such phrases could be construed as claims of priority, which the editors cannot assess and hence must rule out.

Answer 2

I'd say, if your approach is novel there is no problem in mentioning it. A word like "revolutionary" I would be much more careful with. If something causes a revolution or not remains to be seen after publication and feels to me like unnecessarily hyping your results.

So, as long as you use factual words (novel, extends, applies to new application setting) there is no problem in my opinion. But stay away from hype words mostly used by shady business men trying to sell you snake oil.

Answer 3

For "novel", it depends on context. Obviously, any research paper is expected to contain something that is novel, so pointing this out is likely to be redundant. However, many research papers also contain much that is not particularly novel. For example, you might use the well-known techniques A, B, C, D and E to get to a situation where your novel technique F applies. Given the large amount of non-novel material that might end up in the paper, explaining how to use techniques A-E in this particular setting, I don't see any harm in pointing out that technique F is the novel part. On the other hand, you might not necessarily use the word "novel" to say that: you could, instead, write something like "We introduce technique F" or "Our contribution is to use technique F". Or it might even be that the novel part was just using techniques A-E in combination.

But certainly steer clear of describing your own work as "revolutionary". Almost nothing manages to start a revolution before it's even been published. If it truly is revolutionary, this will be obvious, so you don't need to say it; if it turns out not to be revolutionary, you'd look like a dick if you said it was.

I'd also avoid claiming that something has "never been done before" without at least a "to the best of our knowledge"-type disclaimer. At least if a claim to novelty turns out to be incorrect, it's still true that it was novel to you; a claim of "never before" is absolute and, if it's wrong, it's just wrong.

Answer 4

tl;dr: The answer to the titular question (which is asked the opposite way of the questions in the text) is: No, absolutely not. Do use "novel" to highlight what aspects of your paper are new.

This is my perception from a CS field:

Ideally any research paper should contain novel thoughts, ideas, procedures, observations, methods, etc. of some form or another.

Very true. A research paper should contain a contribution that will usually build upon and be embedded within related work.

That being said, is it really necessary to remind the reader about it by using the words novel, revolutionary, or never been done before as opposed to just comparing it to existing studies and noting the differences between the two?

Yes, it is! The contribution, the novel aspects, need to be clearly and explicitly pointed out several times throughout the paper. It is a way of highlighting the point of the paper.

While to you as an author, the contribution is entirely clear - you know exactly what you have done yourself - the same cannot be taken for granted for a reviewer. A reviewer will start reading your paper usually with only a vague grasp of what was there before (reviewers tend to know the particular subfield, but not necessarily your concrete research question for the paper) and starting to build a mental picture of what they perceive as new in your paper.

Therefore, the clearer your contribution is marked as such, the better. Otherwise, you increase the risk that your paper will be rejected simply on the finding that "the contribution is not clear". Explicitly highlighting the contribution involves direct statements such as "Our contributions are ...", but also certain marker words, with "novel" certainly being a frequently used and understood one.

Note that while "novel" is typically understood as such a marker, and while "never been done before" should usually be extended by "to our knowledge" (as opposed to "novel", it is verbose anyway, so three more words won't hurt), I would advise against "revolutionary". The other two options just state something inherent to your contribution, which you can make a statement about, whereas "revolutionary" implies that not only your contribution is novel, but that it changes the way other people will continue to work on the topic. That is a somewhat presumptuous claim to make before your contribution has even been peer-reviewed, let alone published.

Answer 5

You definitely should make clear what the central point of the paper is, but claiming that it is "novel" or something such isn't the right way to do it. You really can't know if you aren't rediscovering something Euler did, and has since then been slumbering in some hitherto unpublished manuscript. Better stick to the facts. You can claim that your method is easier to apply, or simpler to understand, or has wider applicability, or gives better results than competing ones, and stating that conveys useful information. A clam of novelty in itself doesn't.

Answer 6

Highlight and separate the novel parts

I would consider this usage acceptable and recommended not for describing the whole of your result (the novelty of it is implicit, and should be shown during the paper, not just claimed), but in cases where it is neccessary to highlight to the reader (and editors) which part of a nontrivial process is your novel contribution.

For example, if your paper claims that you can achieve great results in frobnicating foobars by doing A, B, C and then D; and A, B and D have often been used in previous approaches (and are still neccessary for your approach) but doing C in this situation is something new - then by all means you should point out explicitly by saying that C in this context is novel, unlike the other parts.

Answer 7

The crux of this debate comes down to the scientific discipline and the definition of 'true'.

Whilst in Mathematics a concept can be said to be provably true or false, very little if anything in Biology is so certain, since we sample observable processes rather than deal in hypothetical absolutes. Our theory of Evolution, although it has never ever been seen to be incorrect, cannot be proven also to be correct - at least not with absolute certainty - because it is a phenomena that is happening all the time, rather than, say, set theory which is demonstrably provable with just a pen and paper today, tomorrow, and in a million years from now.

So when a Biologist says 'X is true', they are always implying 'to the best of my knowledge, X is true'; but to actually add this qualifier to the end of every sentence would be a complete waste of everybody's time because you would have to say it after every point you make.

To say 'you should not use the terminology "novel" because you cannot be certain of novelty' is akin to say 'you should not say polarbears cannot survive in space because you haven't put all the polar bears up there.' OK, I could say "to the best of my knowledge, polarbears cannot live in space", but not only is it a waste of time because it is implied, it is also dangerous because when someone else says "all crabs can live in space", you might entertain the idea, just for a second, that the qualifier "to the best of my knowledge" doesnt apply.

And this, leads to believing nonsense.

The qualifier always applies, no matter who is saying it, and no matter how novel they think it is.