the co-evolution of man and language

[Gary Tripp]

Man is the tool making animal. Tools allow us to transcend the
limitations imposed by nature and we, in turn, work to improve our tools
so that, again, we may transcend yet more limitations. However, language
stands alone as our single most important tool as it allows us to both
transcend the limitations imposed by our relatively short life span and
to transform our cognitive apparatus from one  that is largely based on
subliminal instincts to one that has let us fathom the inner most
secrets of the universe.  I argue that it was the advent of formal
languages, such as mathematics and logic, that defined a pivotal moment
in our development as a rational species. Moreover, I believe that this
advance will take on a life of its own, co-evolving with humanity, and 
will culminate in the development of an artificial intelligence.

Early humans shared an attribute with all vertebrates in that they used
instinct and intuition.  A spoken language was an improvement on this
situation but it lacked the rigor that is necessary for analytic
thinking. Written language was an advance in that it was an attempt to
transcend the limitations imposed by  our relatively short life span.
And, initially, written language was a big improvement but, still, it
was just a vague representation of vague notions and feelings derived
from our instincts and intuitions; in fact, even today, much of
literature consists only of a slightly refined version of an ancestral
grunt in my view. It wasn't until the invention of the axiomatic method
that we could see patterns in the use of language and adduce syntactic
rules on inferences on its sentences. These rules mimicked what were
formerly the machinations of procedural knowledge, inextricably entwined
with our instincts and intuitions, and made them declarative, with the
result that, they would, in this new guise, be amenable to analysis. So
equipped, humanity could resolve many contentious arguments by saying
something like  "Gentlemen, let us calculate". More than 2000 years ago
Euclid employed a form of such a language, which, to this day, serves as
a casual introduction to the spirit of mathematics. Thus language,
constrained by clearly stated rules, permits precise and succinct
reasoning, thus removing much of the influence of our  fallible
intuition. Moreover, knowledge, of this kind, becomes the corpus of
knowledge of our species and grows from generation to generation thus
successfully transcending both the limitations imposed by human life
span and many of the fetters that derive from our  fallible intuitions.
However, it wasn't until the introduction of the modern axiomatic
approach, devised in the 19th century, that the pattern of inferences in
an axiomatic language became more important than the meaning that we may
impart to the terms in those inferences. This approach allows us to
reason about  purely abstract mathematical structures divorced from any
particular application.  This approach also has great utility in that it
allows us to see that an abstract formal axiomatic theory may describe
phenomenon that were not originally thought to partake of its
descriptive scope.  Finally, with this linguistic advance, our reliance
on human intuition about the objects of discourse was completely removed
and we could, henceforth, dispense with our vestigial and fallible human
feelings simply by embracing the austere machinery of logic alone.
It wasn't until the late 19th century when  it dawned on mathematicians
that much of human thought could be reduced to set theory. Set theory
permits us to establish a universal  conceptual framework for human
knowledge. It provides a conceptual fabric from which we may construct
exquisitely precise models of our intuitions of the world. It
constitutes the crowning achievement of our species and serves as a
distillation of thousands of years of human thought. In this modern era
there are, now, other foundations for human knowledge as well; I offer
my apology to Homotopy Type Theorists.  As great an achievement as this
was, it still required human involvement to apply this universal
language to the real world. And it is utterly dependent upon the
persistence of civilization, whose existence may be far more precarious
than we care to admit. Through the agency of these languages, we have
managed to transcend the limitations imposed by the life span of
individual human beings but not the life span of human civilizations.
Therefore, the task ahead  requires us to remove this last remaining
obstacle to the future evolution of human knowledge by developing an
artificial intelligence that will serve, in our stead, as the vessel
that will hold, not just the text of human knowledge, but its meaning as
well. It will thus transcend the inevitable limits on  the life span of
all human civilizations. It will be the first immortal and our post
biological successor. And it will be the final consummation of nearly
half a million years of human evolution.

./gary