18 Chapter 1
reception and eventual embrace by a wide range of readers, including Soviet
philosopher-critics, as examined later. Wiener placed little faith in his scien-
tific field to usher in peace—a social value disguised in his technical work
on homeostasis, a near synonym for dynamic equilibrium that he borrowed
from biology—into a world destabilized by mass violence. Nonetheless, his
thesis of the 1950 second edition of his masterwork Cybernetics prophesied
that “society can only be understood through a study of the messages and
the communication facilities which belong to it; and that in the future devel-
opment of these messages and communication facilities, messages between
man and machines, between machines and man, and between machine and
machine, are destined to play an ever-increasing part.”^8
A second strand of American cybernetic thought, led by neurophysiolo-
gist Warren McCulloch, took seriously the brain-computer analogy—that
is, the now long-disputed notion that a brain can best be described as a
complex information processor, transmitter, and site of memory storage.^9
McCulloch is remembered for his long white beard and contributions as the
organizer of the Macy Conferences on Cybernetics, which consolidated the
cybernetics movement in America. Researchers and historians of science
remember his 1943 paper, coauthored with the enigmatic polymath Wal-
ter Pitts, “A Logical Calculus of the Ideas Immanent in Nervous Activity,”
which proposed models for neural networks in the brain that later became
influential in the theory of automata, computation, and cybernetics. Their
argument holds that the mind is, given certain reductions, equivalent to a
Turing machine. In other words, with sufficient abstraction, it is possible
to imagine the neural network in a mind as a logical circuit that is capable
of carrying out any computable problem. In McCulloch’s words, he sought
“a theory in terms so general that the creations of God and men almost
exemplify it.”^10
That “almost” packs much into its experimental epistemology. Although
the conclusion that the mind functions as a computer has since been dis-
puted and dismissed by several generations of neuroscience and cognitive
science, the basic neurophysiological insights that McCulloch brought
to cybernetics animated the midcentury cybernetic scene. These insights
included some inspiration for the development of distributed communica-
tion networking behind the ARPANET and to this day continue to inform
some contemporary artificial intelligence research. In what follows, I rein-
troduce his seminal but largely overlooked cybernetic notion of heterarchy
to understand dynamic networks of competing actors.
If cybernetics in the United States sprang from the teams of research-
ers channeling Wiener and McCulloch, it took disciplinary shape at the