Evolution as Translation Project
Report: The "Elements" of Web Schemas and Denis
Diderot's Éléments de Physiologie.
This is an outline for a software project that would have
coincided with my translation of Diderot's rare work of
speculative biology, Éléments de Physiologie, on which
I began a translation in 2006.
Not simply describing a software
project, this outline seeks to maintain Diderot's radical
project against classification systems such as those of
Linnaeus, while it intends in turn to locate them along with
formal systems in general, as objects within a continuity of
My translation of Diderot's Éléments de
Physiologie can function as experimental work in rendering
English from French by treating Diderot's text like the very
same empirical objects in his "science of things", a science
first established with the author's Pensées sur
l'interpretation de la nature (1754).
More importantly, my development of a software
architecture for this translation can parallel, through
emergent qualities of software ontologies able to "make
inferences", evolutionary combinations of components behaving
differently upon inhabiting different natural and somatic
Although, while I plan to author its architecture, I
would like to suggest that larger social and cultural
structurings folded even into Nature, will “write” this
architecture as culture/nature, in keeping with Diderot's
notion of emergent development.
Additionally, a conscious diminishing of the
importance of authorship, in poised reference to Diderot's
elimination of final causes of philosophy, will redefine
evolutionary software "localized" to a specific point in our
One question is, what is the place of the formal
system, either linguistic, mathematical, or computational,
when Diderot, according to Pierre Saint-Amand, would have been
highly opposed to all abstract systems except the
infinitesimal calculus of Leibniz?
According to Saint-Amand, Diderot's project of
establishing a science of things in contradistinction to
abstract mathematics, as in Pensées sur
l'interpretation, is strongly opposed, in its emphasis on
nuanced meaning created from experience, to utopian systems,
including Linnaean classification.
The infinitesimal calculus resists Diderot's
criticisms because it conforms to his notion of an
infinite number of beings contained by the concept of one
overall being. Such a being is necessary for making
reference to the whole of Nature, and stresses only
qualitative lives of living beings beyond quantitative
lives fixed into abstract enumerations.
In invoking the infinitesimal calculus, our world
of objects, whether as an idea or as a recourse to an
understanding of this mathematics, is introduced into such
quantitative frames of reference.
If abstract mathematical systems are super-layers
above empirical thinking because they are in fact its poor
shadows and because they are reifications, then the
objects of mathematics - making what is abstract concrete
- are conversely material for empiricists.
Artifacts of mathematics are texts that become
exploratory fields for new empirical conceits.
Artifacts of formal systems, on the other hand,
can be artifacts also of more metaphorical or literary
As modular creations, they are not predetermined
as artifacts of one field but may emerge into any variable
field once within that field.
Similarly computer systems more easily find their
power when they attempt to develop ideas relationally and
According to Bruno Latour and Geneviève Teil in Mechanical
Bodies, Computational Minds (2004), computers should serve in
applications for which they excel: not in emulating formal
systems historically created by humans, but in the manner of a
Hume Machine, an experimental, "skeptical" point of
view on software that they discuss, which maps out sets of
relations symmetrically, relationally, and relativistically.
Latour and Teil remark that in having computers
imitate an aspect of a humanity that separates nature and
society, these machines fall short of any approximation of
meaningful knowledge production that humans have exercised
Because computers are good at determining literal
equivalences between things, they should be used to show
researchers of Latour's concept of "symmetrical
anthropology" relational aspects of their critical topics,
in the manner of the software concept-mapping application, Candide.
Formal systems could at this moment, be left to
Indeed, a formal computational system could be seen
as an expressive use of language by humanity/nature. Formal
systems always refer to things in the world and are coupled
among its actors and fortuitously arranged in our most
infamous binary oppositions.
Formal systems do not transcend a science of things
but say different things in being used within this science of
things, as objects themselves.
More so than Diderot is believed to have thought,
formal systems outside of his texts exist within the
continuity of nature as if they were the things classified by
Linnaeus, rather than the apparatus of classification itself.
The infinitesimal calculus, Diderot's notion of a
chain of being, classification in general, and common concepts
of a “world out there” place natural science's notions of
general and particular into paradox, showing that utopian
classification systems are incorporated into an infinite chain
of beings once presumed to be "real" nature.
Formal systems are contained by rather than coexist
with global notions of a "whole" as demonstrated in
Diderot's Rêve de d'Alembert: readers will likely
never erase from memory the image Diderot gives of mixing
stone with humus and using it to grow vegetables that are
then consumed by animals including humans. As Saint-Amand
has brought to light, this image nearly destroys all systems
of classification by creating a thorough mélange of
animal, vegetable, and mineral realms.
Let us be cognizant of this ultimate place of
classification. Even in the system I am proposing, while
classification is a starting point, it ultimately yields to
what is beyond it, namely life practices that disturb a
clear apprehension of knowledge and continue their projects
even if mired in "disorder".
Classification as "humus" then functions in the
Out of the need to name a variety of species,
classification can be seen to be born, that is from
everyday life and the need for being clear in denoting
things in the world. The world of experience creates at
least psychological or visual classification
While rooted in experience, this form of
classification also has a tendency to lead to the
systematization and creation of a hierarchical ordering
that realizes any number of filiations of the particular
in the general or global.
At the Same time, the infinitesimal calculus,
while potentially describing an infinity of beings,
creates an abstraction from this infinity, in the form of
Together with this "One being" is "la chaîne" or
"continuity" as defined by Diderot, a multiplicity of
beings from an abstraction of one single natural being.
This incorporation of the formal system into objects
of the universe is mirrored in the language of Diderot in the
Éléments de Physiologie especially, his use of “le
tout” and “la partie”.
It is mirrored in the notion of "sensibilité",
described by Diderot as "a quality belonging to the animal,
which forewarns it of relations between itself and all that
Some form of classification now incorporated into the
continuity of living beings might then be seen not to fix
objects, but merely to qualitatively register them in the mind
of its classifiers, enabling a Leibnizian infinitesimally
encoded nature as "One being", in order to permit objects to
emerge from a fabric of things.
We may thus share Diderot's and the 18th century
philosophes' critique of "system" while formalizing a system
attached to a language text and translation, by directing its
neighboring objects within a project of reflection on what we
While an artifact of 18th century science such as the
Éléments de Physiologie can be meta-formatted
according to the conventions of computer science, with such
reflection, the formal system it evokes will be subordinate to
Diderot's text, critical of such systems.
Thus a computer science meta-structuring of Éléments
inhabits the domain of biology and computer science is an
inscription process of Nature.
A semantic specification for a translated Éléments
de Physiologie would thus be one giant whole/part homology
dually transposed upon an act of language translation and
(A closer look parts of the translated Éléments
de Physiologie formatted semantically)