made,do not develop automatically in this way.They can develop only
in the context of a hierarchical structure,which is created by adding one
unit to another and then a third to the combination of the first two.We
have no reason to suppose that in protolanguage any such operation
took place.
In natural language,a variety of operations can be carried out that
involve moving particular constituents around.For instance,instead of,
“Mary baked a cake”you can say,“It was a cake that Mary baked,”or
“A cake was baked by Mary.”These operations involve selecting just the
right constituents—you can’t say,“It was baked a cake that Mary”—and
selecting the right constituents predicates hierarchical structure,since
only items dominated by a single node (and usually by a particular type
of node) can be treated in this way.Moreover,the results of such oper-
ations always change emphasis,and sometimes even meaning,in pre-
cisely predictable ways.In protolanguage,however,anything can be
moved around quite freely,yet apparently without making a difference
in meaning or emphasis,and certainly without making a predictable
difference.
Now,the difference between flat structure (beads on a wire) and hier-
archical structure is absolute,like the difference between life and death,
or married and single,not graded.One cannot be partly married and a
system cannot be partly hierarchical.Somehow a hierarchical system had
to be imposed on protolanguage in a single operation,or else something
else had to be imposed that automatically imposed hierarchical
structure.
To discuss such issues in greater depth would take us too far from the
topic of this chapter.Interested readers will find a full account in Calvin
and Bickerton (in press).For now it is sufficient to note that biomusi-
cology should not jump to the conclusion that the features of music nec-
essarily evolved gradually and were selected for over a long period of
time,the time during which music as we know it today was slowly devel-
oping.Some features may indeed have evolved in this way;others may
not,and it is an empirical question which did and which did not.
This is a crucial point that can hardly be overemphasized.To date,
gradualism seems not to play any significant role in studies of the evo-
lution of music.However,this may merely reflect an early stage of
inquiry,and may result from relative lack of exposure to evolutionary
concerns,rather than from greater sophistication.
Certainly,indifference to evolutionists and their norms characterized
studies of the evolution of language in the previous century.It may there-
fore be the case that,as biomusicology comes farther into the main-
stream of evolutionary studies,it will be infected by the doctrinaire,
quasi-religious gradualism so widespread in evolutionary circles.159 Biomusicology and Language Evolution Studies