27 July 2008

Biosemiotics and symbols

It's been quiet in this space for awhile but this rather long and complex post should make up for that! Here goes:

Biosemiotically speaking, sense perception is a fundamental form of semiosis. But if a sense experience is a sign of some external object, what kind of sign is it—icon, index or symbol? It seems obvious, at first, that a visual image is an icon of its object. But if we take a closer look, especially at a perceptual process which is not visual, it becomes clear that the relation between sign (as sense image) and object (as external reality) is not that simple.
It makes intuitive sense that the perception of high-frequency sound requires receptors that convert sound waves into neural energy. These are the hair cells of the auditory apparatus, which respond to high-frequency sound with a correspondingly high rate of firing. Similarly, hair cells respond with a low rate of firing when presented with low-frequency sound.
Llinás (2001, 219)

The hair cells code sound waves iconically. But as the coded message is transmitted through various stages to the auditory cortex, with one burst of neural activity triggering the next at every stage, it loses in translation its iconic relation to the initial ‘stimulus’.
This tells us something very important: it is not the code or message coming from the outside world that is being transmitted, but rather it is the neuronal element that responds to the message from the outside that is itself the message!
— Llinás (same page)

In other words, the neural activity as interpretant of external events impinging on the senses is now a sign which in turn determines further interpretation. As the semiotic sequence becomes less iconic with respect to the external events, it becomes more relevant to the system's habits, and thus more symbolic. By the time it contributes to a conscious experience, the signal has become a symbol filling a niche in meaning space.

Habits linking neural signals with responses are symbolic, incorporating both iconic and indexical signs into self-organizing systems guided by the mapping of internal models onto pragmatic situations into which behavioral patterns are interwoven. Since this mapping is an outgrowth of intentionality, it is habitual and symbolic: for symbols ‘represent their objects, independently alike of any resemblance or any real connection, because dispositions or factitious habits of their interpreters insure their being so understood’ (Peirce, EP2:460-461, 1909). Though the indexical component of perception is necessary to the system's structural coupling with the external world, the symbolic component is essential to the system's autonomy—that is, to its being a system in the first place.

The main difference between human minds and those of other animals is the extent to which humans make deliberate use of symbols—which we can do because our symbol use can be decoupled from immediate situations. However, this does not mean that symbol use is restricted to humans; the semiotic continuity between human and other lives is unbroken. Indeed the genetic code itself is symbolic. For Baldwin's Dictionary, Peirce wrote that a symbol ‘is constituted a sign merely or mainly by the fact that it is used and understood as such, whether the habit is natural or conventional, and without regard to the motives which originally governed its selection.’ The genome is a sign of the genotype because it is interpreted as such by the system incorporating it.

It seems to follow from this that habits of symbol use can be established by natural selection, given a regular system of sign production, interpretation and replication. Such a system is incorporated in every organism, and in every cell of multicellular organisms. The facts that the genome can be modified by natural selection, and that its expression changes with circumstances in ways that usually turn out to be appropriate, show that it is symbolic, ‘constituted a sign mainly by the fact that it is used as such’ in the course of development. Learning is a similar process within the nervous system, parallel to the evolutionary process (as Bateson pointed out) but at a much faster time scale.

This interpretation of ‘symbolic’ seems at first incompatible with views which take symbol use to be the exclusive province of humans—for instance, Deacon (1997) and Jablonka and Lamb (2005). But this may be due to the habit of taking all symbols to be conventional, whereas for Peirce this was not the case. In a 1904 letter to Welby, he wrote:
I define a Symbol as a sign which is determined by its dynamic object only in the sense that it will be so interpreted. It thus depends either upon a convention, a habit, or a natural disposition of its interpretant, or of the field of its interpretant (that of which the interpretant is a determination).
— PW, 33 (1904)

The ‘field’ here could be taken as a domain or ‘space’ such as a lexicon, and ‘determination’ as selecting or locating an item in that domain. In perception, any given object is perceived to the extent that it determines selection of some pattern from the experiential repertoire of the perceiver (i.e. her Umwelt). Neurodynamically, the pattern can be represented as an attractor in the space of brain activity. These patterns no doubt overlap, accompany or associate with each other in much the same way as the common words in a lexicon. The analogy between the genetic code and language is even more obvious; researchers in genetics are constantly speaking of ‘transcription’, ‘translation’, ‘reading’ and so on, without thinking of their usage as metaphorical. From this it seems clear enough that symbols pervade not only human life but all forms of life which employ the genetic code.

12 July 2008

The testimony of scripture

Lately i've been dipping into Baldwin's Dictionary of Philosophy and Psychology (1901-2) in connection with my ongoing study of Charles S. Peirce and his semiotic. I noticed that an entry on ‘Testimony’, co-authored by Peirce and Baldwin, contains a very concise and cogent set of hermeneutic principles—in other words it gives some very sound guidance on critical reading of scripture—under the guise of a comment on the logic of considering testimony as evidence. I think it pretty well speaks for itself, so here it is:

There is a general tendency to believe what one is told; and, as in the case of other such tendencies, it should at first be followed, although cautiously and tentatively. Even when experience is wanting, as for example in examining a prisoner, although greater caution is required, the proper course is to begin with the presumption that the testimony is true, for unless we make such a presumption, no truth can ever be discovered. It is true that the unlikelihood of the matter of the testimony may cause immediate distrust, or even disbelief of it, but no persons are so frequently deceived as those who stop to weigh likelihoods before accepting or rejecting testimony, and who then form a confident opinion pro or con. Testimony should almost always be accepted as approximately correct, but always strictly on probation, as a subject of examination. In our legal proceedings, witnesses are subject to cross-examination. Everybody is agreed that this is an essential step in the inquiry, but in a historical inquiry no such thing is possible. Still the testimony can be tested in various ways; and this must be done. But in any case, the rendering of the testimony is a fact which needs to be accounted for; and by whatever theory it be proposed to account for it, that theory needs to be checked and tested. Properly handled, false testimony may often yield a great deal of information.

An experimental test may be readily performed by considering the least antecedently likely but necessary or highly probable consequence of the theory, which is susceptible of being confronted with observation direct or indirect. If this consequence is found, notwithstanding its unlikelihood, to be true, there is then some reason for believing in the theory proposed to account for the testimony.

The complete entry (including the first paragraph, omitted here) is on my Peirce-Baldwin page.