40

I'm an undergraduate writing a paper for a philosophy course in logic. My paper is exploring a relationship between Leibniz' logic and graph theory. In writing the paper, I've found some material in Leibniz' work on combinatorics that (to me) strongly resembles graph theory. However, I can't find anything that seriously discusses Leibniz in relation to the history of graph theory --- everything seems to start with Euler. (Most read "Leibniz made early contributions to topology. Later on, Euler developed graph theory" and continue onward from there)

I'd like to be able to say "I looked around on Google scholar, etc, and I couldn't find anything about this topic" by way of suggesting "It's possible that the midterm paper that you're reading is bumping up against an original area of research."

What kind of evidence is appropriate in this situation? Would providing something like the number of results of a search on Google scholar be worthwhile? In general, is there a technique to demonstrate the absence of literature about a certain topic?

5
  • 22
    I looked around on Google scholar, etc — I think/hope you misspelled "library".
    – JeffE
    Commented May 6, 2014 at 9:35
  • 33
    @JeffE why? it's very likely that Google Scholar has more papers and books indexed than the library and the search function works probably better as well.
    – Trylks
    Commented May 6, 2014 at 15:48
  • @JeffE I mostly mean "my school's tool which searches the library and a good chunk of the literature that I have access to," but I'm not sure if such a thing is universal. Commented May 6, 2014 at 16:12
  • 6
    @Trylks You're joking, right? You really expect Google Scholar to include a searchable index of 19th and early 20th century philosophy, mathematics, and scientific history books describing Leibniz's contributions to graph theory?
    – JeffE
    Commented May 7, 2014 at 2:50
  • 27
    @JeffE hahaha! yeah, that's ridiculous!
    – Trylks
    Commented May 7, 2014 at 10:16

6 Answers 6

53

(disclaimer - I come from computer science, and the little I know about conventions in philosophy is from hearsay)

I'd like to be able to say "I looked around on Google scholar, etc, and I couldn't find anything about this topic"

The usual expression for this kind of thing in my field is "to the best of our knowledge, this topic/idea has so far not been considered in literature". It tells the reader that you think that nobody has published this idea yet, but clearly the absence of existing literature is something very difficult to be sure about. Hence the rather defensive formulation.

In general, is there a technique to demonstrate the absence of literature about a certain topic?

Nope. How would you do that? It's a bit like proving a hypothesis - you can only find counterexamples (papers that do cover the topic), but you cannot prove that in some weird journal nobody reads somebody has expressed your thoughts already.

What kind of evidence is appropriate in this situation? Would providing something like the number of results of a search on Google scholar be worthwhile?

The accepted convention in my field is to explain the methodology of how you searched for literature (you had a defined methodology, right?), and then explain / list / discuss what you found. Be specific - which keywords did you use and why? Why Google Scholar and not something else? Can you be sure that all papers that you would expect to be relevant are indexed by Scholar (some of those should be pretty old, long before the Internet was a thing)? Did you also check your library for dead-tree literature? Number of results and or similar metrics seem pretty much useless to me and would likely not improve your paper in any way.

6
  • 23
    +1 for "To the best of our knowledge, this topic/idea has so far not been considered in literature". I am also from Computer Science and have seen and used very similar statements in published papers.
    – wsaleem
    Commented May 6, 2014 at 7:40
  • 2
    This seems like good advice. If you want to make a serious best-reasonable-effort attempt to look for such literature, I recommend asking your subject librarian's advice - only searching Google Scholar will almost certainly be considered inadequate :-)
    – Flyto
    Commented May 6, 2014 at 8:17
  • 4
    +1 for "To the best of our knowledge". But make sure that you do your research. It would be a blow to your face if the paper that you claimed non-existent actually exists in some quite well-known journals. :)
    – justhalf
    Commented May 6, 2014 at 9:15
  • 8
    +1 for "To the best of our knowledge..." but make sure the best of your knowledge actually represents a concerted effort, rather than "We couldn't be bothered to look very hard".
    – Fomite
    Commented May 6, 2014 at 14:43
  • 7
    @wsaleem (and xLeitix) "... not been covered in the literature." To say that something has "not been covered in literature" means that nobody has written a novel about it. Commented May 6, 2014 at 16:44
21

The most relevant works we found are (cite the publication where the water first have been mentioned when your publication is about the invention of the submarine) however they (and say why actually not too relevant).

This will not protect from the deserved criticism if the relevant publications do exist.

4
  • 1
    I am not sure why this has received so few upvotes yet. While I have seen the wording suggested in the answer by xLeitix used in quite some papers, I have also seen the strategy suggested here in use. I think this one shows even better that the author was not "lazy" (of course, only if the claim that there is no literature about the exact topic yet is true), but instead thought thoroughly enough about the topic to point out non-obvious connections to existing works. In addition, space is usually sparse in papers, so every saved word counts. With the "To the best of our knowledge ..." ... Commented May 6, 2014 at 14:52
  • ... version, you use up a whole line of text without really providing any productive information (it's like saying "here is no information"). With this solution, you use the same line of text, but actually provide informative links to remotely related topics, which may inspire readers, or, if nothing else, make your work more easily findable in full-text search. Commented May 6, 2014 at 14:53
  • @O.R.Mapper How is describing some related work that you know is not actually very relevant, and describing why it is not relevant, shorter than writing "To the best of our knowledge"?
    – xLeitix
    Commented May 6, 2014 at 15:00
  • 2
    @xLeitix: The answer says "not too relevant" (emphasis by myself), it does not say "not relevant [at all]". It may not be shorter, but about just as long as the "To the best of our knowledge" sentence, but instead of that sentence actually contains some possibly useful information. Commented May 6, 2014 at 15:04
11

This is not a direct answer to your actual question. Such has been well provided above. Rather, I humbly suggest, as someone who majored in philosophy and has been burned repeatedly by the notion of originality, that when you are tempted to think "it's possible that the midterm paper that you're [writing] is bumping up against an original area of research," refresh yourself with a jaunt to your school's library. There you might find such gems as Benoit B. Mandelbrot's The Fractal Geometry of Nature (San Francisco: W.H. Freeman, 1982) and The Philosophy of Leibniz: Metaphysics and Language by Benson Mates (New York: Oxford University Press, 1989), both of which treat, at least briefly, Leibniz' historical contribution to the development of graph theory. Your library may also have less widely distributed journals and "unpublished" works which are not yet well cataloged on them interwebs.

You might also find that some good scholarly work is done in introductions to collections of original writings. Leroy E Loemker's intro to Philosophical Papers and Letters: A Selection, Gottfried Wilhelm Leibniz (Dordrecht, Holland; Boston: D. Reidel Pub. Co., 1976, 1969) is one such which just so happens to mention your topic. You might also find inspiration in proximal works on such topics as Turing, computational logic, machine theory, and linguistics, to say nothing of the huge body of literature dealing with the interrelationship of analysis situs, graph theory, topology, and calculus.

At the very least, the ability to reference prior works that broach your subject will lend legitimacy to what you might build upon them.

If you're really interested in this subject, though, your best bet is probably to check your syllabus for your professor's office hours, and then drop by for a few minutes to briefly explain your topic, along with your difficulty finding references, and ask her if she has any suggestions as to where you might turn. But I wouldn't mention google.

4
  • Purely as a point of interest -- I the idea which I believe myself to be "bumping up against" is so tenuously-supported that I'm not shocked that I can't find anything (at least in English, my impression is that most serious work on Leibniz is in German). My main source is Couturat's "Logique de Leibniz" -- Couturat describes a part of Leibniz's combinatorics as being inspired by earlier work of Ramon Llull, who calculated (9 choose 9) by counting the number of lines he could draw between pairs of 9 items (which is exactly the complete graph on 9 vertices)... Commented May 7, 2014 at 11:51
  • My paper is a trivial proof that propositions in Leibniz' logic can be encoded in a certain kind of graph. My background reads something like "Here is the status of Leibniz' work among some modern authors. The proof that I am about to provide you probably bears no relation to anything that Leibniz thought, since it relies on math that he did not have. Interestingly, however, this passage of Couturat {quote...} points towards some primitive graph-theoretic notions in relation to logic, suggesting that Leibniz may have had some ideas in this direction."... Commented May 7, 2014 at 11:56
  • (and it now reads) "However, I was unable to find any literature which discussed a link between Leibniz, this technique, and the early history of graph theory, so I am unable to evaluate the merits of this claim." I can't imagine there is any deep philosophical truth to be uncovered there, but I think it's a noteworthy remark in a throwaway undergrad midterm paper. Commented May 7, 2014 at 12:00
  • If I may be so bold, you might be better served to cite something as a jumping off point, even if only to then note that 'there seems to be little discussion of this link' or 'further development'...'in the existing literature' or some such, and then proceed to evaluate the merits yourself, salting your analysis with scraps by recognized scholars here and there when and if you can. I'm not suggesting you twist or stretch others' arguments, rather, deferentially imply credit for planting the seeds. Also, a sentence reading "...I was unable...I am unable...." implies your own failing.
    – Dutch Jeff
    Commented May 7, 2014 at 21:46
3

I phrased a similar idea in a published paper (in a computer science conference) like this:

Although it is well known that space-filling curves can be applied to the problem of approximate nearest neighbor searching, we are not aware of any extension of space-filling curves to approximate range reporting.

2

I actually wrote something similar in a recent paper. I wrote "The author is not aware of any work that is directly comparable to the current paper." I'm not suggesting this is the best phrasing.

After some further details, I then added some specific publications. "Of the many publications relating to x, two representative papers with roughly similar aims to this paper are x1 and x2."

2

I know my answer is late, but this is for those who are recently looking into this topic:

You can claim that you couldn't find enough relevant literature, although, you need to provide proof. The best method used in this situation is systematic literature review (SLR). Using this method you should record everything about your search process such as keywords used, years covered, libraries included, inclusion/exclusion criteria, statistics of results and selected studies. You should record everything, keep in mind that it should be repeatable, meaning that if anyone tries to follow the steps that you recorded he/she will get the same outcomes. Using SLR you make sure that your claim is backed up with evidence, this way you can defend your claim with numbers and facts.

Edit: You can search for Systematic Literature Review or SLR online and you may find a lot of sources teaching you how to conduct it and the required steps.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .