What are we doing?
We will be using a theory of experience that has evolved in language to reconstrue certain biological phenomena.
Why use a theory of experience evolved in natural language?
On the SFL model, scientific theories are higher-level semiotics realised in registers of language and the semiotic systems that language makes possible (Halliday & Matthiessen 1999: 605-6).
And not forgetting … Orgel's Second Rule: Evolution is cleverer than you are.
The main ideational tool we will be using is the identifying figure of being — of the elaborating type. This is the type of figure that is congruently realised in the grammar as an intensive relational clause.
Two relevant examples are:
expression realises content
signifier signifies signified
So, why are we using a figure, rather than, say, some other order of phenomenal complexity such as the element or sequence? Because the key to the construal of experience is the perception of change, and the grammar — the central processing unit of language [Halliday & Matthiessen (2004: 21)] — construes a quantum of change as a figure [Halliday & Matthiessen (1999: 213)], which is congruently realised by a clause — the central processing unit of the grammar [Halliday & Matthiessen (2004: 10)].
And, unlike an element, for example, ‘a figure embodies both analysis and synthesis of our experience of the world: an analysis into component parts, and a synthesis of these parts into a configuration’ [Halliday & Matthiessen (1999: 165)].
And why are we using figures of being, rather than some other type of figure? Because ‘figures of doing and being can be interpreted as complementary perspectives on a quantum of change’ [Halliday & Matthiessen (1999: 132)], and here we are specifically interested in how phenomena are related to each other.
And figures of being, in general, involve not only the experiential relations of identity and ascription, but also the logical relations of expansion, namely: elaboration, extension & enhancement — the relations that also hold between figures in that other order of phenomena: the sequence.
And why are we using elaborating figures of identity, rather than extending or enhancing types ?
Because in this paper we are chiefly interested in the elaborative dimension of material and semiotic complexification, rather than, say, the extending relation of composition , or the enhancing relations of cause or condition, for example.
So with this in mind, the paper is organised into two parts with the same basic 3-part structure.
What differs is the scale of the biological organisation under consideration.
Where the first section is concerned with the evolutionary extension of semiosis beyond the organism, creating social organisation, the second section is concerned with the evolutionary extension of semiosis beyond the cell, creating multicellular organisation.
…where we will consider
first: Intra-organism Semiosis
then: Exteriorised Intra-organism Semiosis
finishing with: Inter-organism Semiosis — specifically: protolanguage.
Now, in considering Intra-organism Semiosis, semiosis in the body, we will be focusing on certain functions of the nervous system and interpreting them in semiotic terms.
Slide 16: The Evolutionary Emergence Of Neurological Systems: Cnidarians (Anenomes, Corals & Jellyfish) & Ctenophores (Comb Jellies)
Organised nervous systems — and so complex active behaviour — first appeared in Cnidaria, the phylum that includes present day marine anemones, corals and jellyfish, and Ctenophora, the phylum that includes present-day comb jellies. For comparison, sponges (Porifora) still lack even the most primitive nervous system, though it was just recently discovered that they have most of the genetic requirements for developing synapses.
The model of brain function we will be briefly using is Gerald Edeman’s Theory Of Neuronal Group Selection. Edelman took Donald Hebb’s 1949 principle of ‘neurons that fire together, wire together’ and interpreted it in selectionist — that is: Darwinian — terms.
For present purposes, we don’t need to go into the depths of the theory, and we can just focus on one function: perceptual categorisation.
On Edelman’s (1992: 87, 125) model, perceptual categorisation involves the strengthening and weakening of synapses within neuronal groups — groups of nerve cells — in two or more functionally different (but connected) maps of neuronal groups, where each independently receiving signals from the outside world — that is, signals from sensory sheets and organs.
For example, when sensory sheets, photoreceptor cells such as the rods and cones of the retina in vertebrates, are activated, as by the impact of photons, the neuronal groups across different maps in the visual cortex that happen to be randomly firing at that time have their synapses strengthened (by the firing), and are then more likely to fire again as a discriminating combination than other neuronal groups (those that were not firing) when sensory sheets are activated again in future. These specific variants in the general population of neuronal groups have been selected in the Darwinian sense of reflecting statistical changes in populations, though here the population is of neuronal groups rather than genes.
This is, of course, oversimplifying considerably, and ignores crucial aspects such as:
the dynamics of the process, and what Edelman calls the value systems that bias the synaptic weighting process,
(Edelman’s ‘value’ refers to settings that were selected as being of adaptive value to ancestors)
but the important point here for us is that perceptual categorisation involves a correlation between specific activations of sensory sheets and activations of specific combinations of neuronal groups.
Visual perception, as in primates for example, involves a correlation between different activations of photoreceptors and different activations of specific combinations of neuronal groups in the visual cortex.
Now, we can interpret this correlation in semiotic terms as a relation of identity.
That is, the material impacts of photons on the retina betoken neuronal values in the visual cortex (and beyond).
This is the brain as recognition system. An event in one domain specifies an event in a second domain.
But to this we need to add the identity relation between the neuronal activations and the perceptual categorisations they realise.
Here we can say that the material impacts of photons on the retina betoken perceptual values realised in neuronal tokens in the visual cortex (and beyond).
Now, stepping back for a moment, we can see that this is the first stage in the explanation of what Bernstein termed ‘how the “outside” becomes the “inside”’.
It can also be seen as the beginning of an expansion of the point made by Halliday & Matthiessen (1999: 609) that:
… what is being construed by the brain is not the environment as such, but the impact of that environment on the organism and the ongoing material and semiotic exchange between the two.
You’ll notice that here we have both material doing, photons impacting retinas, and relational being, the identifying of retinal activations with neuronal activations.
You may have noticed also that most models of brain function tend to conceptualise only in terms of material processes, and doing so typically requires that “information” — in the sense of (pre-existing) “categories of the world” — travel into the brain and then along neural pathways into regions where they are somehow “processed”.
The view we offer here is more in line with Bateson’s view of information (or better: experience) as ‘difference that makes a difference’, and Halliday & Matthiessen’s view that experience is construed as meaning in semiotic systems.
Now we can take this a step further by considering the two directions of elaborating figures of identity: encoding and decoding.
As you remember from Halliday and Matthiessen (2004: 230), in the identifying clauses that construe these figures:
either the Value is encoded or the Token is decoded.
What varies is which participant functions as the Identifier and which functions as the Identified.
In an encoding clause, the Token conflates with the Identifier and the Value with the Identified.
As Halliday & Matthiessen (2004: 230) put it: ‘The identity encodes the Value by reference to the Token’.
So in these two examples, we can say:
The identity encodes the content by reference to the expression.
The identity encodes the signified by reference to the signifier.
Contrariwise, in a decoding clause, the Token conflates with the Identified and the Value with the Identifier.
In this case, again quoting Halliday & Matthiessen (2004: 230): ‘The identity decodes the Token by reference to the Value’
So in these two examples, we can say:
The identity decodes the expression by reference to the content.
The identity decodes the signifier by reference to the signified.
With these distinctions in mind, we are now in a position to construe perceptual categorisation as the encoding and decoding of experience.
So, considering first how perceptual categorisations are encoded in this identity relation,
we can see, in the horizontal relation, that
the identity encodes the perceptual categorisation realised in neuronal groups by reference to the material impact on sensory sheets.
OK, so turning now to how material impacts on sensory sheets are decoded in this identity relation,
we can see, in the horizontal relation, that
the identity decodes the material impact on sensory sheets by reference to the perceptual categorisation realised in neuronal groups.
Well, we have so far modelled one function of the brain in semiotic terms, relating material tokens to perceptual values, but how do we reason from symbolic identity to sensing; that is: from symbolising processes to conscious processing?
Well, the model of experience that has naturally evolved in the English language, as construed by Systemic-Functional Linguistic theory, provides a clue as to how to proceed, and it can be found in the ergative model as it relates to identifying process clauses and mental process clauses. Let’s start with identifying clauses.
As you can see from the slide, in identifying clauses, the ergative function of the participants varies according to the direction of the coding.
The constant is the Identified whose ergative function is Medium (the entity through which the identifying process unfolds), whether the clause is encoding (when it is Token) or decoding (when it is Value).
However, the ergative function of the Identifier is Agent (the external cause of the identifying process) when the clause is encoding (and it is Token), but Range (the domain delimiting the boundary of the identifying process) when the clause is decoding (and it is Value).
[Halliday (1994: 166-7) & Halliday & Matthiessen (2004: 292)]
And in this diagram (after Halliday & Matthiessen 2004: 296 Figure 5.38) we can see the ergative functions of identifying clauses represented
as a nucleus of Process + Medium/Identified,
with its additional participants of Agent/Identifier in encoding clauses,
and Range/Identifier in decoding clauses.
Let’s now turn our attention to mental process clauses.
As you can see from the slide, in mental clauses also, the ergative function of the participants again varies — in this case according to the direction of the sensing: whether it is impinging (the ‘please’ type) or emanating (the ‘like’ type).
Again, there is a constant. in this case it is the Senser whose ergative function is Medium (the entity through which the sensing process unfolds), whether the clause is impinging or emanating.
However, the ergative function of the Phenomenon is Agent (the external cause of the sensing process) when the clause is impinging, but Range (the domain delimiting the boundary of the sensing process) when the clause is emanating.
And in this diagram (after Halliday & Matthiessen 2004: 296 Figure 5.38) we can see the ergative functions of mental clauses represented
as a nucleus of Process + Medium/Senser,
with its additional participants of Agent/Phenomenon in impinging clauses,
and Range/Phenomenon in emanating clauses.
So, the question is: why should the theory of experience that naturally evolved in the English language come up with the property of bi–directionality in just these two process types, identifying and mental?
Well, the simplest explanation of a common property is a common origin, and if identifying relations precede sensing, then the simplest explanation — the one with the fewest additional assumptions — is that mental processes emerge from identifying processes. The diagram on the next slide suggests how that this might happen.
So our “to boldly go” claim is that the Senser emerges as the Medium through which the process of ‘identifying Identifieds’ unfolds; and therefore that the process of sensing — of perception at least — is the process of ‘identifying Identifieds’.
[By the way, if anyone is interested in the origin of other types of sensing, on this model, we can come back to it in Question Time.]
And with the emergence of this “Meta-Medium”, the Senser, and “Meta-Process”, sensing, the valeur of the additional participant, Agent or Range, changes from Identifier of the identifying process to the Phenomenon of the sensing process.
the valeur of the Agent changes from Identifier of an encoding identifying process to the Phenomenon of an impinging sensing process; and
the valeur of the Range changes from Identifier of a decoding identifying process to the Phenomenon of an emanating sensing process.
Note that valeur of ‘perceptual categorisation’ varies according to whether it is construed as the Value assigned to a material impact or a Phenomenon (emanating) that is sensed.
And perhaps it’s of philosophical interest that, on this model:
the Agent of a Senser’ perceiving — its external cause — is the exterior material impact on sensory sheets, as of photons on the retina, whereas
the Range of a Senser’s perceiving — the domain delimiting its boundary — is the interior perceptual categorisation realised in neuronal groups.
OK, so up to this point, we’ve modelled semiosis in the body: from identifying to sensing. Now we turn to a consideration of the external manifestations of these internal processes.
So, with the preceding discussion in mind [!], let’s begin with the (interior) identity relation between a ‘senser-sensing’ and the neurological form in which this medium-process (function) is realised.
Now, various types of emoting, desiring, thinking and perceiving are expressed by perceivable actions of the body. For example,
hearing by ear-pricking (gazelle)
curiosity by gazing intently (octopus)
desire by lip-licking (hyæna)
anger by very loud harsh noise-making (rainbow lorikeet)
The last of these makes clear that the doing includes the Range of the doing across perceptual modalities, so as to potentially include, for example, sounds or smells (pheromones).
In such cases, where bodily action manifests states of consciousness, there is an identity relation between an interior Value, what a senser is sensing — realised in neurological system-&-process — and an exterior Token, what a body is doing.