35The fact is that at this stage of linguistic inquiry, almost every new language that comes
under the microscope reveals unanticipated new features (p. 432).
The number of languages for which adequate ethnobiological studies have
been published is likely to be vanishingly small in comparison (e.g. the survey of
39 languages in [ 113 ]), and the task of formulating meaningful generalizations
correspondingly more diffi cult. The fact that unrelated languages from many dif-
ferent parts of the world behave in a similar fashion is interesting, and perhaps
indicative of a deeper phenomenon worthy of investigation, but the smaller the
initial sample size, the more likely that the observed similarity will turn out to be
a trend, and not a ‘universal’.
The task of challenging an absolute ‘universal’ is an easy one. Strictly speaking,
all it would take is a single, verifi able counterexample to bring the entire edifi ce of
the ‘universal’ come crashing down. For after all, what are ‘universals’, if not test-
able hypotheses? The person seeking to prolong the life of a ‘universal’, on the other
hand, has the unenviable task of having to defend the idea in question against a
potentially unforeseen piece of empirical data. Two options are available to such a
person: address the challenge directly, and argue why it is not a valid refutation of
the ‘universal’, or modify the ‘universal’ (i.e. the theory) in accordance with the
new data. It is not a valid strategy to point out the existence of published studies that
support the ‘universal’ as defence against the new data. Those studies are the foun-
dation that the ‘universal’ was built upon, and the introduction of new data that do
not conform to the predictions made by the ‘universal’ becomes an issue that has to
be dealt with in its own right.
2.4 Problems with Berlin ’s Ethnobiological Classifi cation
Many of the general principles proposed by Berlin in his 1992 monograph have
resonated with ethnobiologists working in different parts of the world, as is evi-
denced by the many scholarly publications purporting to lend support to certain
‘universals’. The following discussion (along with further tests of Berlin’s predic-
tions later in the book) is not meant to suggest that Berlin’s general principles will
never apply to any language. On the contrary, I believe that the general principles
likely capture some important truths about how languages will often converge on
certain solutions in spite of being unrelated. What is being argued here is that these
general principles are by no means universal (i.e. there will be languages that violate
one or more of the general principles), and this is made abundantly clear by the data
presented in Chaps. 3 and 4. A more immediate concern of this section is the theo-
retical justifi cation underlying the existence of general principles in the fi rst place.
There may well be identifi able reasons that explain why many languages behave in
similar ways with regard to ethnoclassifi cation and nomenclature , but I do not
believe that they are identifi ed by Berlin.
2.4 Problems with Berlin’s Ethnobiological Classifi cation