evidence of causation, and causation

ww.hardconsonant.org/?p=30″>dwelling on this, but it was a really interesting series of posts.  So maybe she’s getting what she wanted?)

Anyway, here’s my situation.  I’m studying land pressures, with an eye toward how they produce conflict.  There are historical reasons for land pressures here: namely large-scale displacement, dispossession, and now return.  And there are structural reasons for land pressures: limited land, high population growth, poor productivity.  And there are institutional reasons: legal confusion and (my interest) state allocation of remaining public lands to political allies rather than landless residents.

Among these various causes, the situation of returnees gets by far the most attention.  However, returnees do not factor greatly in the actual incidence of land conflict.  Land disputes clog the courts; but returnees are parties to only something like 10% of these.  This is a higher-than-representative figure, but hardly dominating.  Other surveys of land conflicts outside of courts have similar results.

When I have discussed this with people, they reply that returnees may not be a large percentage of the land conflicts but that those conflicts are at the greatest risk of triggering political instability.  It is true that land disputes have fueled political conflict in the past.  But current land-related violence reported in the press is more likely to describe conflict over succession than over returns.

So I’m wondering what to do with this “evidence”.  The apocryphal taxi driver wisdom (in this case supported by law professors, human rights activists, etc.) says that the land crisis is about returns.  The data says that it is about something else.  The apocryphal taxi drivers respond that the concern is about something uncountable — the likelihood a small but identifiable class of property disputes will trigger future political conflict.  And the evidence replies….?

One the one hand, this could be a classic case of evidence-triumphalism.  People’s views are clouded by subjectivity.  Data shows us the right path.  I should feel empowered to drive ahead with alternative theories.

On the other hand, maybe this is a classic case of data being beside the point.  Data as the lamp-post in the dark, where people search for their lost keys not because the keys are there but because they can’t look anywhere else.  The right answer is not found from random sampling of measurable events, but from asking the right questions to the right people, who can produce information based on experience and exposure.  I had argued, in response to TiA’s posts, that there are valid epistemologies aside from sociological methods.  It seems that the consensus of local actors would be one of those.

For the time being, I’m probably going to do the one thing that everyone knows shouldn’t be done: selectively choose my evidence to support my case.  The numbers support alternative theories, so I’m going to use that as enough justification for continuing to work on alternative theories for now.  At some point, though, I’m going to have to confront conventional wisdom.

Whatever.  I’m a lawyer.  Cherry-picking evidence is my job.