There is no definitive distinction between papers and articles that can be applied to all scientific disciplines. Usage varies between disciplines. and within disciplines it can vary depending on context.
Both the examples quoted refer to ‘writings’ that are surveys (in other areas often termed reviews) — one in the field of a social science (economics) and the other in a numerical science (computing). However the term science is also (and perhaps more) associated with the experimental sciences (physics, chemistry and biology), where the types of ‘writings’ are different and where different words are used to distinguish them.
Articles and papers in the Experimental Sciences
Let me illustrate this for the Biomolecular Sciences (biochemistry, molecular biology, molecular genetics and the like). As a practitioner in this area, when I hear these terms, e.g. talking to colleagues, I understand:
Paper: A report of a piece of experimental research work in which the original data presented by the authors was central to interpretation and conclusions regarding advancement of knowledge and understanding of the field.
Article: A review or commentary in which the author was discussing the previously published work of others (perhaps including his own) in attempting to provide a perspective of the field or to present a new theory/model/interpretation by integrating such work.
However, despite this professional conversational use of the terms, if I go to any specific journal — here the US heavyweight, Journal of Biological Chemistry (JBC) — I would find a somewhat different usage:
JBC publishes several types of articles but only two of those can be submitted as an unsolicited manuscript: regular papers and accelerated communications.
Thus, JBC regards all the ‘writings’ it publishes as ‘articles’, in common with other journals such as The Journal of Biophysics , and this is consistent with general non-scientific usage — “I read an article in the Financial Times yesterday…”
The way JBC uses ‘regular paper’, is consistent with my specialist conversational definition (above), and although it doesn’t actually say what types of ‘article’ are unsolicited, but if you look at a table of contents of the journal, you would conclude that for this journal it is ‘minireviews’ and historical appraisals of the work of individual scientists.
The Journal of Biophysics only uses the term ‘paper’ in describing only one of its categories of ‘article’:
Comments to the Editor | Short commentaries on a paper published earlier in BJ.
Again using ‘paper’ rather in the sense I defined above.
To conclude, in the extended sense used by peer-reviewed journals in the experimental sciences, all published ‘papers’ can be referred to as articles, but not all articles would be referred to as ‘papers’. (One wouldn’t use ‘paper’ for an editorial, a news item and generally not for a review.) This is exactly the opposite conclusion reached by @1006a from his reading of the OED.
Conflict with the OED and non-experimental sciences
How can one resolve the conflict with the OED, mentioned above? I think the OED describes more traditional usage in the non-experimental sciences and the arts. It is pertinent, in this respect, to consider the phrase “reading a paper”.
As far as my area of science goes, this is just a rather outdated term for presenting one’s results orally at a conference. The talk in itself is transitory, the abstract unreviewed, and the information conveyed will most probably be published elsewhere.
However for colleagues in computing science the talk is likely to be based on a ‘paper’ that has been submitted to the conference organisers, selected after peer-review, and will be published as conference proceedings or in a journal associated with the conference. This is more in line with traditional non-scientific academic presentations, although in this case the ‘paper’ might never have been published.
The difference would seem to derive in part from whether the field of science is one in which original work is in the form of ideas or in the form of measurements and their interpretation.
Q. I plan to submit a research article to Nature, but I am not sure what is difference between Nature article and letter. I notice there is usually only one article per issue of Nature, but several letters. I do not know how to decide to submit a manuscript as an article format or as a letter format. Would you please explain this to me?
A. Articles are original reports whose conclusions represent a substantial advance in understanding of an important problem and have immediate, far-reaching implications.
Letters are short reports of original research focused on an outstanding finding whose importance means that it will be of interest to scientists in other fields.
More details are available at the journal’s guide to authors and at the Nature Publishing Group author and reviewers’ website, where there is a description of the article types published in all the Nature journals.