Molecular Ethology, An Immodest Proposal for Semantic Clarification - Understanding Understanding: Essays on Cybernetics and Cognition - page 158

III. Biophysics

A. General Remarks
The question now arises whether or not one can identify structural or func-
tional units in living organisms which can be interpreted in terms of the
purely mathematical objects mentioned previously, the “tiles,” the
“automata,” the “finite function machines,” etc. This method of approach,
first making an interpretation and then looking for confirming entities,
seems to run counter to “the scientific method” in which the “facts” are sup-
posed to precede their interpretation. However, what is reported as “fact”
has gone through the observer's cognitive system which provides him, so
to say, with a priori interpretations. Since our business here is to identify
the mechanisms that observe observers (i.e., becoming “self-observers”),
we are justified in postulating first the necessary functional structure of
these mechanisms. Moreover, this is indeed a popular approach, as seen by
the frequent use of terms like “trace,”“engram,”“store,”“read-in,”“read-
out,” etc., when mechanisms of memory are discussed. Clearly, here too the
metaphor precedes the observations. But metaphors have in common with
interpretations the quality of being neither true nor false; they are only
useful, useless, or misleading.
When a functional unit is conceptually isolated—an animal ,a brain ,the
cerebellum , neural nuclei ,a single neuron ,a synapse ,a cell , the organelles ,
the genomes , and other molecular building blocks—in its abstract sense the
concept of “machine” applied to these units is useful, if it were only to dis-
cipline the user of this concept to identify properly the structural and func-
tional components of his “machine.” Indeed, the notions of the finite state
machine, or all its methodological relatives, have contributed—explicitly or
implicitly—much to the understanding of a large variety of such functional
units. For instance, the utility of the concepts “transcript,”“en-coding,”“de-
coding,”“computation,” etc., in molecular genetics cannot be denied.
Let the n- sequence of the four bases ( b = 4) of a particular DNA mole-
cule be represented by a

- be
an operation which transforms the symbols (0, 1, 2, 3) Æ (3, 2, 1, / ), in that
order, with 0 ∫ thymine, 1 ∫ cytosine, 2 ∫ guanine, 3 ∫ adenine, and / ∫ uracil,
and I be the identity operation I (

-number ( n , b ) [see Eq. (33)]; let Tr ( ) =

- ( n , b )] =

( n /3, a ) =
m( m , a ), with a = 20, and j = 0,1,...,19,representing the 20 amino acids
of the polypeptide chain. Then

) = ; finally, let F[

(i) DNA replication: = I ( ) (55a)

- = Tr (

(ii) DNA/RNA transcript: ) (55b)

- )

(iii) Protein synthesis: m=F( (55c)

While the operations I and Tr require only trivial machines for the
process of transcription, F is a recursive computation of the form