Soils and landscapes are subject to historical and spatial contingency, lea
ding to locally unique pedologic features. This can make broad-scale genera
lizations difficult, impractical, or even impossible. The development of ve
rtical texture-contrast soils with argillic horizons, for example, is poten
tially subject to all general forms of spatial and historical contingency.
A case study in east Texas shows evidence both supporting and refuting five
general classes of explanation for the formation of vertical textural cont
rasts. Multiple causality is Likely, and attempts to apply any single expla
nation to a county-size area land sometimes to a pedon) are not likely to b
e successful. The implication is not that pedologists should abandon the se
arch for generalizations, but that the context in which laws and generaliza
tions are developed needs rethinking. Explanatory constructs should be form
ulated not with the notion that a single explanation is likely to be applic
able to most soils, but with the idea that multiple causality and polygenes
is are likely, and that location-specific characteristics cannot be ignored
. The search is directed not toward a single principle to explain the major
ity of cases and against which exceptions can be judged, but toward a set o
f principles that define the possibilities (or probabilities). Two analogie
s that may be useful in addressing historical and spatial contingency in so
ils are proposed, based on demographic and synoptic metaphors. (C) 2001 Els
evier Science B.V. All rights reserved.