The analysis in Nature Ecology & Evolution is here: http://go.nature.com/2hBd0yU
On the door of one of our experimental rooms in Orsay (France), a small flyer cheekily claims “A month in the laboratory can often save an hour in the library”. Walking daily passed that door for years, this quote from famous chemist Frank Westheimer has come to persuade us that one can never read too much science.
But in the Information Age, the superabundance of peer-reviewed articles — around 1.5 million now published each year — together with the annoying failure of physicists to invent time compression, mean that we need to select the relatively few papers we have the time to read.
Yet, the average researcher is not lazy. Studies have established that we read a rather surprisingly large number of papers each year (check out our article to find out whether your reading intensity is below average). But as scientists tend to become ever-more specialized, so too does the scope of our readings. This tapering allows us to get to grips with the known, and dig deeper into the unknown; however, the associated risk is to become overly narrow in one’s knowledge base, and to miss useful papers that do not fall directly into our limited field of expertise. This is especially relevant in a rich and varied discipline such as ecology.
We thus decided to construct a list of ‘must read’ papers that ideally, every young scientist in ecology should read, regardless of their field. Obviously, establishing the criterion of ‘importance’ was a nightmarish task to avoid becoming overly subjective, but there was one way to make this at least partially meaningful and tractable.
We achieved this simply by relying on the experience of our peers, and on the weight of their collective opinions. We asked 665 editorial members of the most authoritative journals in general ecology to provide us with three to five papers that they would nominate for such a list. Later, we asked them to vote for the papers they collectively nominated. Since experienced scientists are used to evaluating a paper’s importance, collectively they ought to come up with as few biases as possible.
We list the resulting 100 ‘must read’ papers in the article we published in Nature Ecology and Evolution, together with a basic analysis of the age, field, and type of these papers. This list comes with its share of surprises, both in terms of which papers emerged from the selection process, and of correlations — or lack thereof — with journal impact factor, article citation rates, and other elements.
For example, it would probably surprise many that the 100 most-recommended articles in ecology have a median age of almost 40 years. Indeed, one of the reviewers of our article — an expert in communication sciences — was initially convinced that this had to be a mistake, because in this expert’s opinion, the most recommended papers should be no more than just a few years old.
The content of this must-read list and its characteristics also teach us a lot about our science, which is deeply embedded in concepts and theory, but is also unexpectedly well-balanced among many sub-disciplines.
As we often say to our students, “ … your study doesn’t exist until it is published”. In retrospect, we should probably now also update this to “… your study doesn’t exist until it is read”. So, to make the most of science, and to make science exist, one must read, read, read, and read some more.