Our thinking in this chapter makes a significant shift away from received opinion, and into territory that will be unfamiliar to most readers. Some of its implications will easily be missed on first reading. It is a distillation of several different sources - all intellectually challenging - but the power and scope of this new way of understanding make it well worth the effort it will take.
One of the shifts I am making is an important re-application of the principle of natural selection. We are not concerned here with the evolution of species (where, as I have argued elsewhere(1), the principle of natural selection may have been seriously misapplied) but with the evolution of behaviour in the here-and-now. I am arguing that natural selection - the principle that declares that serious mistakes in the conduct of life get automatically deleted from the hereditary line - is something akin to a universal constraint on living processes. It defines for every creature a behavioural range within which variations can occur whilst life is able to continue. It is the discipline to which all life is subject, not merely on pain of death, but of the extinction of the germ-line (i.e. the whole series of potential descendants). Though we do not know exactly how the living cell organises itself so as to satisfy this constraint, it clearly manages to do so by some means or another.
In this chapter we are looking at what a living creature does within the range of behaviour allowed it by the law of natural selection. There is a repertoire of patterns of self-adjustment that enable the creature in question to stay alive and eventually to reproduce itself. We are seeking to understand that particular kind of intelligence that we human beings have in common with the single-celled organisms. This is an intelligence not to be looked down upon, since it has evidently made possible the survival of primitive organisms throughout the millennia. Also, if you accept (as I do) that the theory of evolution entails the historical relatedness of all living species then you should accept the consequence: this primitive bio-intelligence has provided the fundamental pattern out of which the entire kingdom of multi-cellular plants and animals has succeeded in evolving.
Biology of Cognition
I want to consider the behaviour of a single-celled animal (such as Amoeba Proteus) from the vantage point of cognitive biology, as it has been developed in the work of Humberto Maturana and Francesco Varela in the late 1960s and early 1970s. These researchers, one a neurophysiologist and the other a biologist and mathematician, invented the concept of auto-poiesis,which they hold to be the defining characteristic of the living creature. It connotes the organised activity which creates the components needed to constitute its own organisation. We should note that this definition captures the closed-cycle nature of the living process. This, argue Maturana and Varela, is what effectively demarcates the boundary between the living and the non-living.
One unique feature of living protoplasm is the manufacture of large, chemically complex, individually sculpted molecules based on long carbon chains of various patterns of construction. These molecules are precisely organised in relation to one another in space and time. They exist in a milieu which also contains a whole variety of simpler chemical components whose concentrations are maintained within a remarkably narrow range (corresponding, I have been told, to the chemical composition of the sea in some primordial era). More impressive, perhaps, is the fact that this precise and detailed regulation is happening in the face of myriad comings and goings. The large molecules (starches, lipids, proteins and nucleic acids) have relatively stable locations, subject to a process of continuing re-modelling and maintenance, but the elemental and simple ionic complexes (sodium, potassium, calcium, chloride, bicarbonate etc) seem to be skeetering about haphazardly, except that their comings and goings through the outer membrane of the organism are regulated by a variety of mechanisms.
Our main focus here
will be two-fold. We are looking within the structure of the organism
for the characteristic patterns of event which are the
precursors of emotional intelligence; and we are looking at the
creature's overall behaviour - its readiness to cope with
life-threatening situations, the apparent foresight with which it
interacts with objects in its surroundings, and its seeming dedication
to keeping its own life processes in balance. All of this evokes in us
the sense that the creature knows what it is doing. More
specifically, I am interested in the relationship between these two
levels - how the processes inside the organism's boundary contribute to
its survival as a whole.
The account which follows has a number of diagrams attached and you need to be forewarned that these are not quite what they seem. It is natural to interpret a continuous, closed line such as we see in figure 1 as representing something in physical space. This is not the case here, because we are not interested in the physical boundary of the organism, as much as in the boundary of its behaviour. These are therefore systems diagrams (albeit very simple ones) which summarise and categorise an enormous number of processes which happen within a given period of time. The inside of the closed figure represents the "domain of all possible internal states of the organism, compatible with its being alive". The line itself represents the demarcation between the states of the organism where it is still viable, and (on the outside of the line) states that are not compatible with the continuation of life. Each point within this boundary represents a determinate state of the living organism. (Clearly I am not addressing any serious structural detail in this diagram - a detailed behavioural map would require the few square centimetres of our diagram to open up like one of those Mandelbrot fractal pictures, in which more and more precise structure emerges the closer we approach it.)
Any process contributing to the life of the organism can be thought of as a transition from one "state" to another "state". This event is represented by a line passing from one point to another, the arrow denoting a forward movement in time. (See figure 2.)
Since this is a lower-level process contributing to the higher-level configuration we call "life", we need to recognize it as something more than a single event. It is a recurrent event whose re-occurrence fulfils a repeating function as one segment of a cyclical series of causes and effects. There are two important points of contrast between this and our common-sense notion of linear causation (which serves us so well in the organisation of our practical life - in such activities as building a house or designing a motor-car). They can best be demonstrated visually, and so we shall represent the common-sense causation as a straight-line series where one event directly influences the next in the chain:-
By contrast, the cyclical patterns characteristic of living processes have come to arrange themselves in such a way that the outcome of any particular physiological state is also a contributing factor in a chain of causation which will lead to a recurrence of that same state - in a future cycle of events.
The second contrast is that these cyclical patterns often intersect with a great many other similar ones - which means that any one event in the making, is subject to a convergence of causes. This makes it impossible for us as observers to say with certainty that event " a1" actually caused event "a2" to occur; but we may be reasonably sure that it entered into the causal ambience out of which "a2" emerged. This is no different from multi-factorial causation in other circumstances - such as a large number of over-laden lorries crossing a bridge already weakened by repeated attacks of frost, and the scheduled re-painting of the bridge (which would have revealed dangerous fracture-lines in one of the supporting arches) being delayed by the bad weather, leading to a major collapse of the bridge.
Certain patterns of circular causation have the additional property which systems engineers call "ultrastability". An ultrastable pattern is one which repeats itself in regular fashion, and is able to absorb the incidental impact of a range of other events, so that its repetitions are maintained within stable limits. All living organisms have this property and thus continuously maintain their own internal balance under a range of stresses. An important aspect of this property, which we might easily overlook, is that the organism need not have any previous acquaintance with - or understanding of - the specific nature of the disturbing event in response to which it is having to restore its own equilibrium. This differs from a typical computer-programmed activity, which requires a complete specification of all relevant circumstances in advance, with precise sets of instructions for how to deal with all contingencies. The auto-poietic system need make no calculations in respect of any world external to itself; it is enough that it measure its own internal states and that it respond according to what it finds within.
One example of this ultrastable cyclical pattern is that of DNA and RNA organising the synthesis of protein which catalyses the process of sugar synthesis and metabolism, which in turn provides the energy to maintain the whole system which replicates DNA.
In his classic study of control and communications patterns applied to both living and mechanical systems, "Decision and Control", Stafford Beer argues that an essential component of the ultra-stable process is the generation of some random variation at certain stages in the process (but contained within the over-arching self-adjusting cycle). This allows the system to find and then to amplify behavioural variants which improve its stability through time. So even at this most basic of levels of organisation, there is something like an exploratory or creative function which can modify and improve the circular causal organisation, without dismantling its essential continuity. This is perhaps the crux of the circular practical logic underpinning the living process - that it is the integrity of the cycle which is the invariant in the system. The individual steps may be adjusted and varied in all kinds of ways, on condition that the main cycle is able to continue uninterrupted.
These basic features need to be taken as read, when we look at "changes of state" within the single-celled organism we are considering at the moment. Thus the single transition a1 >> a2 as represented in figure 4 must be understood as one event within an auto-poietic cycle. We, as observers, have singled it out for our attention as if it began at (a1) and ended at (a2), but it has taken its origin in auto-poietic process and it is leading into further auto-poietic process - if it did not, it would not belong in this diagram.
We now need to make a further distinction, between the range of states we could call "physiologically comfortable" and those we would call "physiologically stressed". The states lying within the zone of stress are deviations from the optimum that are not so great as to be incompatible with life, but great enough to be registered within the system in such a way that they trigger further changes, operating in the direction of restoring the equilibrium back to within the zone of comfort. (See figure 5, below.) The change a1 >> a2 was a movement away from physiological comfort. a2 >> a3 was a restoration of comfort, but to another state (a3) distinct from (a1). Of course this is not meant to represent any specific biochemical pathway of events, but it takes for granted that the living cell is already organised to make these serial and simultaneous biochemical adjustments, and that its success in doing so has been the pre-condition for the million-year long survival of this cell and its ancestral line.
We know that this auto-poietic process depends upon a myriad of interlocking, cyclical biochemical pathways, a great many of which have been explored by molecular biologists and biochemists over the past several decades. Without the full complement of these biochemical processes the auto-poietic cycle could not be maintained - but we do not need to understand these processes in detail here.
This is in exact analogy to the correct notes needed to execute "I Can't Give You Anything But Love!" and the convergence of strategies which achieved the victory of Wellington's army at Waterloo - we recognize the overall result but may be largely ignorant of the details.
The continuous monitoring of internal states, characteristic of the simplest of organisms, is a close analogy with our own experience of feeling. A physiological state which falls in the zone of stress corresponds to a feeling of pain or discomfort. It signals that "something is wrong" and that "something needs to be done". Looking back at our earlier description, we could say that this feeling of stress enters into the overall causal ambience so as to increase the probability that subsequent changes in internal state will move the system closer to the zone of physiological comfort.
In general, the longer, and the more seriously, our creature finds itself outside its physiological comfort-zone, the more strongly its priorities become diverted to the matter of restoring physiological comfort by whatever means are available. We do not have to think of our little friend as being aware of anything outside the domain of its own internal states. It is enough that it registers its own states of physiological stress and comfort, and has appropriate response pathways for maintaining itself through the stresses and strains of its insecure existence.
Another fundamental concept we need to borrow from Maturana and Varela is that of structural coupling. This is the automatic consequence of an auto-poietic creature living within a milieu which contains needed raw materials and potential threats and dangers. The creature is organised so that outside influences are to a great extent buffered by the internally driven ultrastable response patterns. In effect the organism is resonating like an Aeolian harp in response to every impact from outside. The pattern of this response is determined by the organism's own structure; it is not like a direct imprint or impression of the outside world. The effect through time is that the patterns of event within the cell become spontaneously coupled with patterns of event in the surrounding milieu with increasing precision - while their detail remains determined moment by moment (in all important respects) by the structure of the organism
Some details of the coupling will have developed over a span of many past generations; some will have been acquired over this creature's own lifetime, but its actual mechanism always operates in present time, as a consequence of the organism's continuous ultrastable adjustment procedures.
The essential point about structural coupling is subtle, and depends upon our being clear about the difference between our point of view as observers of the Amoeba in its natural medium, and the point of view of the Amoeba itself. As observers we are able to interact with, and to keep track of, both the amoeba and some of the events surrounding it. We are able to recognize that events within the cell are moving in repeating patterns which are effectively tracking the patterns of event in the surrounding world. If we now shift in our imagination to try to consider the creature's own point of view (within the framework of understanding we have been developing here) we may be surprised to realize that it has no need of any "information from outside", nor does it need to know that there is any such thing as an outside. Its own auto-poietic activity (in the context of its structural coupling with the surrounding medium) is giving it all the information it needs, inscribed within the changing pattern of its own internal states.
If we want an analogy with something in our own life experience, this would be a feeling about something which guides our action independently of any need for conscious recognition of what we are doing(2). We easily overlook such responses; our own are by definition unaware, and when we read other people's responses to situations we tend to attribute conscious perception to them just because our own reading has been conscious. Consider the following, however: I am approaching a bus-stop, walking in the direction of traffic so that I see the people waiting and beginning to shift their positions in response to a bus which I cannot see (since it is approaching from behind me). I have several levels of response open to me. The simplest is that I sense the approach of the bus without making any conscious inferences at all. This is the response which is equivalent to the level of structural coupling - the people at the bus-stop are shifting around as if at the approach of a bus, and I take in this ambience by a kind of osmosis, so that I too feel as if there is a bus approaching. The conscious route to this awareness might be that I notice people looking up and starting to move, and I look behind me and see the bus for myself. My assumption now is probably that "they saw the bus first", attributing a conscious perception to the other people. The important insight for us here is that the people would probably start to move anyway - regardless of whether they consciously noticed the bus or not. We are often so much in tune with surrounding events that we come up with sensitive responses without any need for thought.
Returning to our single-celled organism, we see there is no need to attribute any knowledge of the outside world at all - neither conscious nor unconscious - since the relevant behaviours arise out of the sequences of internal state, which are structurally coupled to patterns of events on the outside. We do naturally think of the amoeba as responding to objects in its environment when we watch its approach to food and illumination, and when we notice the violent withdrawal of its pseudopodia (the outward-flowing columns of protoplasm which extend into limb-like processes during its normal movement in and around its milieu) consequent upon physical impact or the presence of acid. But it is also true at the level of the creature's own organisation that it is simply "doing what comes naturally"; anything which impacts upon the organism to induce a discernible change in its internal state will provoke an auto-poietic response - which means a consequent change in state that helps to maintain the system within its zone of physiological comfort.
Rudiments of Language
Our sense that the creature "knows what it is doing" is all the more compelling when we consider the dance which takes place between two amoebae in close proximity. We cannot help but think of their very evident shifts and adjustments as being "calls" and "responses" in relation to each other. We need an extra conventional sign on our diagram (figure 6), which will stand for the shift in creature "A" as it impacts upon the internal dynamics of creature "B". That is the meaning of the star and the projecting barbed line which passes from "A" to "B".
And in the case illustrated here it looks very much as if the change a1 >> a2 has triggered "B" into "becoming upset" (b1 >> b2). This is not necessarily the case. It is also possible that the move from b1 to b2 could have arisen out of some important internal dynamic of "B" - or indeed an interaction of several different factors. Therefore it would not be correct to say that "a1 >> a2" caused the subsequent shift "b1 >> b2". Rather we should say that "a1 >> a2" entered into the overall causal ambience out of which "b1 >> b2" emerged.
At this level of description we are still not claiming any awareness on B's part, of what has happened in A. We are simply saying that "B's" state is significantly different from what it would have been, if a1 >> a2 had not happened. In Maturana and Varela's language, A is not imparting information to B, but is simply a source of "perturbations" (i.e. disturbances). In depicting this in diagram form, we can make the assumption that every change in the state of "A" may have some impact on the subsequent change of state which takes place in "B". ("No response" would thus be a special case of the general situation "there will be a response of some kind".)
We can now move forward to considering the case of serial interaction, in which each shift in one organism acts as a possible trigger for a shift in the other, which in turn acts as a possible trigger for the first.... and so on until they either mate, wander off into separate parts of the pond, or just continue to co-exist in a companionable way. This is schematised in figure 7, below. There is no further mechanism required to achieve this spontaneous process of continuous mutual adjustment between the two creatures. They will learn, in effect, to take advantage of whatever benefits might accrue from their mutual proximity, and to make the best of any incidental disadantages. The relationship will be more or less complex, depending upon the compexity of the coupling between the sequence of states in the one creature, and the sequence of states in the other. (The more "important" they are to each other, the more intense and more precise will be the impacts of the one upon the other.)
So what we are looking at is a process of structural coupling between two organisms which follows automatically from their arrival in each other's neighbourhood. Maturana and Varela point out that this is a new level of complexity in its own right, and gives rise to what they call a consensual domain. The patterns of behaviour in this domain amount to a private language having developed spontaneously between the two participants, and it is therefore also termed a domain of linguistic behaviour. This is really an extraordinary situation for us to contemplate, because most current theories of language assume that there has to be a pre-established code or convention before people can communicate through language. Yet here we have two very simple-minded creatures who seem to be capable of entering into their own linguistic community, establishing whatever codes or conventions they need as they go along!
With respect to our own exploration of facts, feelings and their inter-relationships we need to be clear that these unicellular organisms are communicating exclusively in terms of feelings at the stage I have described. Each message is, in effect: "This is what it feels like for me, to be on the receiving end of what you just did!" However, as Maturana and Varela point out, an exclusively connotational language such as this can still give rise to a subtle and precise mutual orientation. For instance we need very little further elaboration of the theory to provide a satisfactory account of the honey bees' dance - a dance which can orient a swarm of bees to a particular patch of a particular species of flower from several miles away(3). Our only problem is that in order to read the dance we would have to have the inner neurological structure of a honey bee - in which case we might find it very difficult to communicate our findings to our human colleagues!
What we have found here is something resembling a spontaneous intelligence, an automatic process of coming-into-tune, which works at the level of the simplest organic processes. It does not require any consciousness, nor intelligence as we normally understand the word, in order to operate. Our working assumption in the chapters which follow is going to be that this rudimentary kind of intelligence is working busily within us at every moment of our lives, and that it underpins our higher-level faculties in ways which we normally take for granted. The implications of this assumption are extensive, but there are two major challenges entailed in it which I want to take special note of.
Firstly, we need to learn to recognize when and where this organic intelligence can be trusted to do its work, free from unnecessary interference. This is so that the higher levels of our intentional behaviour are enabled to work in harmony with the lower levels, in contrast with that sad but common situation where we frustrate our own best intentions by pushing againstthe deeper flow.
The other challenge is to identify those sensibilities, empathies and rationalities which are most distinctly human, and try to better understand the relationship between these and the organic intelligence which underlies them.
Please Note:- this section begins a sequence of sustained argument, linking the theory of auto-poiesis to the radical personal and cultural enquiry which is the sphere of interest of this web-site overall. The argument continues under the following headings:-
NOTES TO THIS SECTION
2. The importance of "tacit knowledge" in the performance of every human skill is one of the central arguments in Michael Polanyi's authoritative and influential book Human Knowledge. Tacit knowing is what apprentices learn on their long years of training, and something the master craftsman can rarely put into words. This non-conscious level of "knowing what to do" is the exact counterpart of Maturana and Varela's structural coupling.
3. .I would hypothesise that the structural coupling which enables this highly specialised and effective communication has been built up over many billions of years of single-minded practice. The precise nature of the coupling can only be traced by means of painstaking observation and hypothesis, of course.
© all content: copyright reserved, Michael Roth, January 2004