The dictionary definition of a law is: "Generalized formulation based on a
series of events or processes observed to recur regularly under certain con
ditions; a widely observable tendency". I argue that ecology has numerous l
aws in this sense of the word, in the form of widespread, repeatable patter
ns in nature, but hardly any laws that are universally true. Typically, in
other words, ecological patterns and the laws, rules and mechanisms that un
derpin them are contingent on the organisms involved, and their environment
. This contingency is manageable at a relatively simple level of ecological
organisation (for example the population dynamics of single and small numb
ers of species), and re-emerges also in a manageable form in large sets of
species, over large spatial scales, or over long time periods, in the form
of detail-free statistical patterns - recently called 'macroecology'. The c
ontingency becomes overwhelmingly complicated at intermediate scales, chara
cteristic of community ecology; where there are a large number of case hist
ories, and very little other than weak, fuzzy generalisations. These argume
nts are illustrated by focusing on examples of typical studies in community
ecology, and by way of contrast, on the macroecological relationship that
emerges between local species richness and the size of the regional pool of
species. The emergent pattern illustrated by local vs regional richness pl
ots is extremely simple, despite the vast number of contingent processes an
d interactions involved in its generation. To discover general patterns, la
ws and rules in nature, ecology may need to pay less attention to the 'midd
le ground' of community ecology, relying less on reductionism and experimen
tal manipulation. but increasing research efforts into macroecology.