296 Raphael Falk
basis of heredity with Watson and Crick’s model of DNA, further directed research
studies to gene function rather than to embryonic development and differentiation.
Benzer’s detailed mapping of the gene and Yanofsky’s demonstration on the
colinearity of the gene sequence and the that of protein made the elucidation of
the process of gene function more urgent. However, the next step was primar-
ily a task of biochemistry of synthesizing proteins in the test tube [Rheinberger,
1997]. Genetic analysis stepped in again with making sense of the processes of
transcription and translation by providing a model of genetic regulation of gene
action [Jacob and Monod, 1961].
Many believed at the mid-1960s that molecular genetic analysis has exhausted
itself [Stent, 1968], but development and differentiation was not resolved by the
molecular genetic analysis of gene regulation and the paradigm of Jacob and
Monod [1961]. While molecular biologists encountered the complexity of eu-
karyotic gene organization, “classical” genetic analysis managed to improve its
methodology and increasingly to analyze gene function in a system-context, thus
addressing more relevantly the “classical” problems of embryology. It took some
time for the molecular biologists to appreciate Lewis’s breakthrough in provid-
ing a phenomenological model of genetic control of segmental differentiation (see
Morange [2000, 196]), but once obtained, genetic analysis of embryological devel-
opment and differentiation became increasingly molecular.
Evelyn Fox Keller attributes these developments to the change of the mecha-
nistic, lineal mode of thinking to that of the kibernetic, informational feedback
era (Keller [1995; 2000; 2002], and many others) that changed the image of the
gene as anacting agent to that of anactivatedagent. There is no doubt that
the ”informational” metaphor had an important role in genetic thought. It must
however be kept in mind that this metaphor was far from the mathematical-
kibernetic notion of information theory, which dealt with the probabilistic reliabil-
ity of transmission of signs, with no reference to their semantic contents, whereas
the information-metaphor of genetic analysis was crucially dependent on the com-
parison and transmission ofsemanticinformation, as obtained by the methodology
of hybridization, whether that of living organisms or that of DNA and/or RNA
nucleotide sequences. More importantly, Keller ignores the internal developments
in the conceptions and techniques of the sciences involved. Both Development Sys-
tems Analysis (see Oyama [1985]), and the notion of Punctuated Evolution (see
Eldredge and Gould [1972]), which played major roles in the new organismal-look,
were engendered faraway from the foci ofmolecular-biology, by avid opponents
of the information notion. By the time computing algorithms and machines for
analysis of very large number of data became available the information metaphor
was long forgotten in molecular biology.
Keller is right in pointing at Goldschmidt, who “was typically grandiose, leaning
always toward overarching generalization”, unrelentingly insisted on embryological
development as systems of interacting and coordinated reactions: “[H]is search
was precisely for the dynamic properties of such systems (Zusammenspiel der
Reaktionen). To Goldschmidt gene action meant that genes were both catalysts