Some months ago Alexandra wrote about how she had to ‘re-learn’ writing for a blog after spending most of her time writing for academic publications. I am faced with the converse: re-learning how to read academic publications after spending most of my time reading laymen’s writings.
Here is a good example of what I am talking about. It is a sentence found in “On Formally Undecidable Propositions in Principia Mathematica and Related Systems I” by Gödel:
To every w-consistent recursive class k of formulae there correspond recursive class-signs r, such that neither v Gen r nor Neg( v Gen r ) belongs to Flg(k) (where v is the free variable of r).
Ok, this is a fairly specific example and a fair amount of familiarity in number theory is probably required to parse it. What this actually means in normal English though is rather straightforward according to Gödel, Escher, Bach by Hofstadter:
All consistent axiomatic formulations of number theory include undecidable propositions.
Neat and tidy… Or is it?
Well, we seem to have made progress in the sense that we now have the information in a format in which it can be manipulated easily in certain ways e.g., we can generalize the statement to mean ‘a set of things includes a specific subset of things’. This opens to us our accumulated knowledge about set theory and other related things.
On the flip side, it also blurs some definitions: what is a ‘consistent axiomatic formulation of number theory’? That definition was more detailed in the first version provided by Gödel. Even then though, the original statement was itself using terms that required definitions and so a more verbose (but equivalent) statement could be made (and so on).
Reading scientific papers is akin to the above exercise in navigation across different levels of precision and thinking. Our mind has to juggle from very precise to very general statements. This is hard because it is not something a new graduate student is used to.
The specific obstacles to reading I identified are the following (if you have others let me know):
Precision. Scientific texts aim to be as precise as possible. Most findings apply in very specific situation and as such use language to precisely denote those situations. We don’t need that level of precision in our daily lives. I can say: “Monkeys stole our secret stash of bananas yesterday” and my friend will understand me even if the monkeys in questions are really Papio Ursinus.
Conciseness. Scientific texts have a limited size and thus need to synthesize ideas into smaller words to make them fit both on paper and in our minds. Precision tries to maximize details while conciseness tries to minimize length – this often puts them at odds with each other.
‘Obscurantisme Terroriste‘ i.e. obfuscation. Usually this expression is used to denounce works that are deliberately obscure. I don’t think it is common, but it can seem that way. This is probably the worst kind of hurdle to understanding. As as sidenote, one could expect commercial whitepapers to fit in that category for purposes of monetary gains in exchange for further explanations, but my experience has not proven this. In fact research papers coming from commercial entities tend to be much more goal oriented and thus understandable.
The best ways I found to overcome these issues are by reading older papers and by reading popularizing books. Older papers have the advantage of less technical jargon baggage in their explanations. ‘Popular’ science books go through the first level of translation (like we did above) and so give you a starting point on which to continue.
How do you deal with complex papers?