I’ve just been reading some of Tim Berners-Lee’s Weaving the Web about building the World Wide Web and it put me in mind of Paul Frijters’ recent post on teaching the social sciences. Paul argued that:
The biggest change needed is to teach the material in terms of basic patterns, with more complex arguments taught later as combinations of basic patterns. Another change needed is to enforce a single language on the entire curriculum. Finally, what is needed is far more use of virtual reality-teaching and field trips so that students experience the phenomena they are meant to understand, unlocking their visual acuity and emotional skills as learning tools. Students should learn with their whole being, not merely with their abstractive capacities.
Regarding the first two points, 1Berners-Lee’s frustrations with the world before he changed it seem to mirror Paul’s. He was frustrated with the ‘tree like’ organisation of knowledge 2. and wanted to invent a looser, more ‘associative’ form of knowledge architecture which he called ‘mesh’.
Be that as it may this seems like quite a big deal to me. Indeed, as the internet grew in order to effectively scale it needed to move away from ‘tree-like’ architectures for sending data packets through the net as outlined in this write up of the Border Gateway Protocol (or as us aficionados call it the BGP – some of us have only been aficionados for the last couple of minutes, but we’ll leave that to one side).
Anyway, the way knowledge started being classified in the 19th century was by discipline. Sometimes a discipline disappeared because it was discredited – as in the case of phrenology. But mostly the disciplines stayed in place and each spawned endless sub-disciplines. The disciplines are tree-like structures of knowledge, sometimes paying obsessive attention to their unitary structure as in the case of micro-foundations in economics and selfish genes as in the case of neo-Darwinism.
There are occasional cross-overs, as in the case of behavioural science and economics or imaging and psychology or evolutionary psychology for instance. Interdisciplinarity is spoken of at least by some with reverence, but all the disciplinary incentives are against it and the tree-like structures of each discipline remain in place. Thus these crossovers might be likened to rhizomes rather than any real challenge to the tree-like structure of disciplines and plantation like structures of knowledge generally.
As knowledge has proliferated over the last two centuries, you’d expect some re-architecture of the relations between disciplines, but there’s been no systematic change whatever, and what occasional change there has been, has been limited to ad hoc marriages of sub-disciplines where new technical possibilities present themselves or where exhaustion with the sterility of one discipline sets in and some (usually very limited) reaction is in order. An example of the former is where MRI imaging is now used in psychology and even in philosophy and of the latter is behavioural economics or the empirical or ‘Freakonomic’ turn in economics (though this is also to some extent the response to the new technical possibilities of ‘big data’ and desktop computing).
- I won’t comment on his third point other than including it here because of its importance and my strong agreement with it. ↩
- The paper proposing that he work on what became the World Wide Web had a heading “The problem with trees” and began “Many systems are organised hierarchically” ↩