17The inclusion of naturalistic speech is essential when asking “ how ” questions in
ethnobiological research (such as “ How do people normally group together named
organisms? ”), as standard elicitation practices—interviews, naming tasks, sorting
tasks—create artifi cial contexts that have the potential to generate unnatural
responses. The idea that context and basic conversational principles can signifi -
cantly infl uence the interpretation of an utterance has been known to linguists for
quite some time [ 78 ]. Unfortunately, some ethnobiologists who work on folk clas-
sifi cations take their consultants’ responses during elicitation sessions far too liter-
ally, ignoring the fact that many features of such responses would be affected by the
very specifi c contexts of those sessions. Berlin , for instance, has argued that
‘ generic ’ level categories have a special status in folk taxonomies, because they are
the ones generally provided in response to a question regarding an organism’s iden-
tity. As an example, Berlin presents us with the following hypothetical conversation
between an ethnographer and a consultant (adapted from p. 52):
Ethnographer: That (over there), what’s its name?
Consultant: (It’s named a) tree.
Ethnographer: That (over there), what’s its name?
Consultant: (It’s) also (called a) tree.
The above exchange is characterized by Berlin as ‘improbable’, ‘ludicrous’ (and
rightly so), and is meant to support his argument that ‘ generic ’ taxa—the “ names of
common speech ” (p. 53), the types of labels that would seem most appropriate in
this context —would be used preferentially because they label the “ smallest funda-
mental biological discontinuities easily recognized in any particular habitat ” (‘tree’
in this case would belong to Berlin’s ‘ life form ’ rank). I disagree with this conclu-
sion, for Berlin is confounding cognitive salience with some other highly relevant
features of “ common speech ”, namely Grice ’s [ 78 ] maxims of quality (i.e. accu-
racy), quantity (adequacy of information), relevance (making utterances relevant to
the issues currently being discussed) and manner (avoiding ambiguity). First, we
need to assume that the ethnographer and the (willing) consultant are both desirous
of obeying the Cooperative Principle.^3 It can then be safely assumed that since the
ethnographer is a naïve (in the psychological sense) outsider, and since the consul-
tant is aware of this, the latter party will try to accommodate the former’s lack of
knowledge by being maximally helpful and relevant. The consultant will have made
the assumption that the ethnographer is probably familiar with trees, and because
s/he is new to the fi eld site, may choose not to burden him/her with a highly precise
subordinate label. The only option remaining to the consultant—one that is simul-
taneously accurate, adequate, relevant and clear—is to provide a ‘generic’ label as
a compromise. One can easily imagine specifi c scenarios where taxa of ranks other
than ‘generic’ constitute the most appropriate response to an utterance, in keeping
with the Cooperative Principle:
(^3) The Cooperative Principle [ 78 ] states: “Make your contribution such as it is required, at the stage
at which it occurs, by the accepted purpose or direction of the talk exchange in which you are
engaged.”
1.4 Concerns About Scope and Method