John Searle’s book entitled ‘Intentionality’ could otherwise be entitled ‘The principles and conditional characteristics of conceptual representation’. I say this because his irreducible stance on Intentionality is tightly restricted to and fixated on conceptually confined modes of analysis of mental representation expressible as a function within a “circle of Intentional concepts”.
there is no nonintentional standpoint from which we can survey the relations between Intentional states and their conditions of satisfaction. Any analysis must take place from within the circle of Intentional concepts. p.79
Outside of this “circle of Intentional concepts” which he calls “the Network”, all other components of mentality can be considered a part of “the Background”:
The Background is a set of nonrepresentational mental capacities that enable all representing to take place. p.143
Let the “hypothesis of the Background” be the claim that Intentional states are underlain by nonrepresentational, preintentional capacities. p.144
Intentional states only have their conditions of satisfaction, and the whole Intentional Network only functions, against a Background of what I will, for want of a better term, call nonrepresentational mental capacities. p.20
An Intentional state only determines its conditions of satisfaction – and thus only is the state that it is – given its position in a Network of other Intentional states and against a Background of practices and preintentional assumptions that are neither themselves Intentional states nor are they parts of the conditions of satisfaction of Intentional states. p.19
It is worth noting at this point, that when John Searle states that the Background consists of nonrepresentational mental capacities, his definition of representation is itself very restrictive, for in his view,
There is probably no more abused a term in the history of philosophy than “representation”… p.11
The Background is not on the periphery of Intentionality but permeates the entire Network of Intentional states; since without the Background the states could not function… p.151
Given its simplicity, why does Searle’s account of Intentionality come across as complicated? The following quotation may give some clues why this is the case:
In my view it is not possible to give a logical analysis of the Intentionality of the mental in terms of simpler notions, since Intentionality is, so to speak, a ground floor property of the mind, not a logically complex feature built up by combining simpler elements…. Any explanation of Intentionality, therefore, takes place within the circle of Intentional concepts. p.26
In denying explanatory analysis Searle is committed to explaining Intentionality in terms of an isolated non-reductive “family of notions”, namely “Intentional content, psychological mode, conditions of satisfaction, direction of fit, causal self-referentiality, direction of causation, Network, and Background”. In part, the complexities of explanation lie in the problem of articulating the demarcation point between Network and Background, between concepts and non-concepts, between representation and non-representation.
I intend to explain why Searle’s account fails because of a single but vital omission in his thesis. To do this does not entail demoting ‘Intentionality’ in the way that Daniel Dennett appears to suggest through his Intentional Stance. Rather I will show what Searle omits to even consider, namely that Intentionality is reducible to more fundamental levels. This realisation comes at a price however. The price is that we deny Brentano’s (1874) objective which was to seek and to determine laws for only mental phenomena in order to divorce the science of psychology from philosophy and physiology. Searle is loyal to Brentano’s bold endeavour, but I think the boundaries required of it – those of the circle; the Network and the Background – are artificial and ultimately are descriptive rather than explanatory. Even Searle in his online lecture series emphasises that naive realism is counter intuitive (http://www.youtube.com/playlist?list=PL553DCA4DB88B0408).
Read this critique of Searle’s Intentionality as a pdf.
Searle states that “…the key to understanding representation is conditions of satisfaction.” which is fine, but he continues, “Every Intentional state with a direction of fit is a representation of its conditions of satisfaction.” (p.13) Why, “every Intentional state” specifically? Why not rephrase as follows “…the key to understanding representation is conditions of satisfaction. Every
Intentional state with a direction of fit is a representation of its conditions of satisfaction.” p.13 I see no reason why we should stick with Searle’s original statement that requires the qualification of states to be “Intentional”, where the term “Intentional” is as defined by Searle and without qualification. Alternatively, with my innocuous editing, where we can assume that ‘any self-perpetuating state’ may be a subject of process that institutes some form of coherent mechanism and action, we have dispensed with Searle’s mentalist and conceptual presumptions regarding Intentionality and opened out the analysis leaving us with the question whether neural based cognitive operations are the only kind of process that could generate representational and Intentional characteristics.
Searle says the Background is nonrepresentational:
Intentional contents in general and experiences in particular are internally related in a holistic way to other Intentional contents (the Network) and to nonrepresentational capacities (the Background). They are internally related in the sense that they could not have the conditions of satisfaction that they do except in relation to the rest of the Network and the Background. p.66
With this artificial orthodoxy; and its division between Network and Background, Searle omits the possibility that there are forms of representation other than conceptual types: in contrast to Dennett and his exploratory stance on Intentional, Searle is not prepared to consider that nonconceptual mental features may be representational in nature. I have argued elsewhere (for example, in my critique of ‘Ten Problems of Consciousness’), that for anything to be coherently responsive to its environmental conditions is for it to have some mode of relating to its environment. In this way, there must be some aspect to a responsive construction of this kind – specifically in its manner of responding to environment – that, ultimately, makes its construct representational in some way. Searle does not entertain this possibility, rejecting it with the following:
…suppose that all Intentionalistic mental life and all cognitive capacities could be entirely reduced to representations: beliefs, desires, internalized rules, knowledge that such and such is the case, etc. Each of these representations would be expressible as an explicit semantic content (even though, of course, many of them are unconscious and thus unavailable to the introspection of the agent), and mental processes would consist in going from one such semantic content to another. But there are certain difficulties with this picture. The semantic contents with which the conception provides us are not self-applying. Even given the semantic contents we still have to know what to do with them, how to apply them, and that knowledge cannot consist in further semantic contents without infinite regress. p.152
But here, one can see with clarity Searle’s fixation on the notion that the only form of representation is conceptual; with the acceptable argument that semantic contents cannot extend indefinitely. In response, and in order to negate this line of reasoning, one need only point out that there might be ‘some class, or classes of’ representation that is/are neither semantic nor conceptual in construction:
Surely, any Background mental capabilities must be contextual. A Background cannot be isolated from what it is that the Background ‘is of’ or ‘is about’. Does it not make intuitive sense to consider the ‘properties’ of the Background as having evolved in a manner that relates to the environment such that one might say of those properties, that they are representational? After all, different animal species have different background skills, habits, abilities and the such like. These background capabilities must be pertinent and representationally relevant for each species. Given this evolutionary precedent for “the Background” to be pertinent for all species, what is a Background if not in some certain manner representational, in this expansionist view? One need only consider that the Background itself, i.e., the evolved capacities of the Background, is qualitatively relevant to the environment because it, and its representational capacity, is relevant to survival.
A proponent of Searle’s stance on Intentionality may well have a difficulty accepting this notion, but surprisingly, Searle would not, as he reveals in a footnote on p. 141,
Intentional states with a direction of fit have contents which determine their conditions of satisfaction. But they do not function in an independent or atomistic fashion, for each Intentional state has its content and determines its conditions of satisfaction only in relation to numerous other Intentional states. [At this point, John Searle refers the reader to a footnote: “I am discussing human Intentional states such as perception, beliefs, desires, and intentions. Perhaps there might be more biologically primitive Intentional states which do not require a Network, or perhaps not even a Background“. [Italics added] p.141
Consider the following:
There is a primitive organism species. We shall name it Seareal. By primitive, I am saying that it has a limited neural network; much like Nematode. It does not conceptualise, it does not learn from its mistakes or adapt behaviourally on a realtime basis to environmental experiences. Rather, its behaviours are inherited; behavioural responses are innately being predetermined by its genes alone.
It inhabits water and has hairy filaments along its body that are responsive to vibrations. These filaments respond to running water acting as sensors to its vibratory effect. When an excitatory threshold is initiated, innate propulsive activity ensues. It propels itself aimlessly until the currents become reduced, at which point it stops propelling itself.
There is also a predator that feasts on Seareal, much as cows feast on grass.
Over generations, the hairy projections – connected by nerve cells to a network – become increasingly sensitive to variations of vibration amplitude and length, and therefore to current flow and even to current direction and propulsion develops in such manner as benefits survival too. By some quirk of mutation within the network cluster, one member of the species develops an Innate Rapid Propulsion response to sudden accentuated vibration. There is no particular reason why this should have come about. Ordinarily this variation would be of no benefit, but as an incidental consequence, this branch of Seareal finds itself avoiding the once omnipotent predator whose clumsy movements happen to generate sudden accentuated vibrations as it lopes about in the water. Over time, offspring that mirror this new capability begin to populate the environment more readily as they successfully avoid the voracious predator; their new physiology appears to be more responsive to environmental conditions.
On reflection we have good reason for putting faith in the accuracy of the conceived principles articulated by Wallace and Darwin. Adaptation is a qualitative environmentally referential consequence; reflecting the value and the benefit of ‘good’ environmentally responsive and accurate biochemical and biomechanical mechanism. With increasing complexity, not just with regard neurological facility but in relation to all biochemical and biomechanical physiologies, we get increasingly sophisticated innately acquired environmentally responsive constructs.
What can be said of the physiologies and the resulting behaviours of the species Seareal?
1. I would argue that these physiologies represent ‘something’ about the nature of the environment, if only as an accident of mutation. On first discovering Seareal, a scientist might observe these characteristics and use understandings about the creature’s behaviours and its biomechanisms to glean ‘knowledge’ about Seareal’s niche environment. This is only possible because these complex physiologies are representational. How is there any accurate correlation between an organism’s biochemical/ biomechanical physiology and the nature of environmental conditions without some form of representational contextual construct? I concede that there is no thinking-representation, there is no conceptual-representation, and there is no experientially adaptable learning capability. But the physiology of the evolving species – rather than that of the individuals of the species importantly – does relate to the environment. How might this be described if not in terms of representation? These physiological constructs, I deem representational. They are representational because they carry information about the environment with which they have interacted and from which they have adapted over generations.
2. I would also argue that the consequential behaviour does demonstrate Intentionality. This is not to say that an individual Seareal organism ‘believes’ anything about ‘a predator’, considers how it might avoid immanent ‘pain of death’, thinks how it might ‘benefit’ from certain actions, nor that it can ‘reason’ why it should behave in such and such manner: it does not think about its purpose, desires, beliefs, or behaviour. For there is no conceptual realisation. It merely enacts its innately acquired behaviours. But there is Intentionality in the species-specific actions that are enacted over generations as the species seeks an all-embracingly stable physiological adaptation.
3. All aspects of experience are rooted in physiologically qualitative and environmentally relevant representations. This applies as much to visual perceptions as anything else. I submit that Searle’s approach to perception is far less persuasive in two regards.
i) First, he clings to the belief that there is no explanatory expansionist framework for different classes of representation and that there is no representational feature governing the roots of perception. In reference to a yellow station wagon he writes:
Well, what must be the case in the station wagon scene in order that the experience be a veridical one? At least this much: the world must be as it visually seems to me that it is, and furthermore its being that way must be what causes me to have the visual experience which constitutes its seeming to be that way. And it is this combination that I am trying to capture in the representation of the Intentional content [of perceptual experience]. p.48-49
A snake, a bat and a human’s perceptual experience of a yellow station wagon will differ markedly. In each case the nature of the animals’ physiologies is the limiting factor. As a human, I do not smell the car’s metal and rubber, nor see the heat of the engine, nor can I perceive it equally well in night as day. What one can say of all forms of perceptual experience (of which these are three example), is that they are accurate reflections – accurate representations – about the qualitative relevance of a certain world-view. They can be thought of as informational constructs that are accurate because accuracy is essential to the survival of the particular and complex needs of the organism and ultimately its species: all forms of perceptual experience are qualitatively relevant to the survival of the respective organism. False or inaccurate perceptions, sometimes referred to as hallucinations and illusions, are likely to be detrimental to survival. So, whilst “the world must be as it visually seems to me that it is” – as is also true of the snake and the bat – in many sophisticated ways my perceptual experiences inform me but only in a qualitatively relevant fashion with regard the specific survival needs of my human body.
ii) Secondly, Searle is forced to limit his enquiry to conceptual interpretation due to non-reductive assumptions. He is forced to analyse perception merely in terms of an apparatus for manipulation by conceptual representation.
When I stand and look at a car, let us say a yellow station wagon…. How does the seeing work?… I mean how does it work conceptually…? p.37
Visual experiences, like beliefs and desires, are characteristically identified and described [italics added] in terms of their Intentional content. There is no way to give a complete description of my belief without saying [italics added] what it is a belief that and similarly there is not way to describe my visual experience without saying [italics added] what it is an experience of. p.43
Of course, there is nothing wrong in looking in detail at the conceptual and verbally referential and descriptive character of perceptual interpretation and engagement, but the essence of the phenomenon of perception is not conceptual. There is a spectrum, a hierarchy of representation (c.f. the HCT infographic). One must look at the processes of the brain from bottom up to understand how primal and innate affective moods, for example, become integrated with perceptual experience to instantiate qualitative characterisation. There is good reason to believe that colour, movement, and shade should have qualitative characteristics. If, as a hunter gatherer, I seek spherical red berries that contain sugar and other nutrients, then, if not by innate sensory stimulation alone, a primal reward mechanism needs to be in play to incentivise and encourage the seeking of said spherical red objects without distinction. The anatomy of primal neurological reward and punishment mechanisms have been isolated and mapped in all those mammals and birds that have been studied using deep brain stimulation (DBS) techniques. Panksepp, (2011 – Cross-Species Affective Neuroscience Decoding of the Primal Affective Experiences of Humans and Related Animals) labels the identified seven characterised reward and punishment flavours isolated in the Medial Forebrain Bundle (MFB) as having SEEKING, FEAR, LUST, CARE, PANIC (GRIEF), PLAY features, with the most powerful SEEKING area generating moods from intense euphoria to mild enthusiasm. With these primary areas hooked into the learning mechanisms of the cortex, one can easily understand how perceived colours, shapes, and movements, as an example, might instantiate qualitative mood states and affective motivational and avoidance impressions. How these kinds of qualitative characterisations and representations about environmental experience become the focal point of conceptualised and verbalised reasoning is yet another layer of hierarchical complexity parsed in the neocortex.
In contrast, Searle states,
The main difficulty with a representative theory of perception is that the notion of resemblance between the things we perceive, the sense data, and the thing that the sense data represent, the material object, must be unintelligible since the object term is by definition inaccessible to the senses. It is absolutely invisible and otherwise imperceptible. As Berkeley pointed out, it makes no sense to say that the shape and color we see resemble the shape and color of an object which is absolutely invisible or otherwise inaccessible to any of our senses. p.59-60
What I am arguing is that there is every reason – as the neuroanatomy, developmental evolutionary evidence, and commonsense suggests – to suppose that there are different layers of representation that are pertinent to sensory perception (some conscious others subconscious) and that such things as shape and colour are qualitatively represented in entirely distinct ways by primary, secondary, and tertiary processes cf. figures 2 and 3 below: Panksepp’s diagram of nested hierarchies and Pharoah’s diagram of Hierarchical constructs.
These representational layers correspond to innately acquired neurobiological mechanisms, realtime perceptually evaluative mechanisms, and conceptualising capabilities. Each layer informs an organism about the environment (both the external and internal environment) in different kinds of ways. There are different classes of informational construct. Neurological mechanisms have the capability of assessing the qualitative relevance of a changing landscape of experience. For non-human animals, this assimilative, evaluative and prioritising cognitive capacity is enough to institute highly adaptive behaviours. What we do as humans is objectify these phenomenal impressions with conceptual representations, each of which comes to form our symbolised conceptual characterisation about the world in which we experience. Thus we have layers of representation to explain what Berkeley saw only as a gaping void.
What we are faced with is the likelihood – I say likelihood, because evolution of complex form could not be so simple as Searle might hope – that representation and Intentionality are moulded in different ways by different classes or types of mechanism.
…the key to understanding representation is conditions of satisfaction. Every
Inevitably, one comes to the question, what differing ways can states represent aspects of the environment? Alternatively, what are the different classes of mechanism that are representational?
There is every reason to suppose that these classes of representation with their particular brand of Intentionality are constructed in an hierarchical fashion – this is the evolutionary and neuroscientific evidence and now we have the theoretical framework provided by Hierarchical Construct Theory. What is required is qualification through an expansionist classification whereby Searle’s ‘representation’ becomes exclusively ‘The Category of Conceptual-Representation’, contrasting with ‘The Category of Physiological-Representation’. Both types of representation are embodied by a particular type of structure that bears a particular form of informed construct regarding the nature of environment through its specific forms of mechanism and process.
I submit, that as an explanation for “the Background”, this idea of hierarchical layers of different classes of representation implemented by different forms of mechanism is more intuitively reliable that Searle’s explanation that,
…the sense in which the visual experience is self-referential is simply that it figures in its own conditions of satisfaction…. what is seen are objects and states of affairs, and part of the conditions of satisfaction of the visual experience of seeing them is that the experience itself must be caused by what is seen. p.49
‘Mother Nature’ and Intentional States
1. Consider atomic elements/compounds, replicating organisms, animal consciousness, and human awareness, each as examples of distinctive types of construct.
2. A construct is a coherent whole that is conditionally constituted by non-aggregate parts that are continually interacting dynamically.
3. A construct is identified as a coherent whole by virtue of its temporal stability.
4. Constructs that are not stable tend to dissipate, whilst those that are stable tend to predominate.
5. Thus, constructs that are stable tend toward temporal longevity necessarily proliferating at the expense of those that do not.
6. Consequently, one might say that “Mother Nature” is, by default, in the business of proliferation through the maintenance of enhanced temporal stabilities.
7. Indeed, for whatever reason, Newton’s Laws of Motion are fundamentally concerned with this characterisation of stability acquisition and maintenance: All colliding bodies negotiate a compromise toward a stable equilibrium state.
8. Environmental interaction tends to lead to variety of construct forms due to the reconstitution of stability due to destabilising effect of interaction.
A. I argue, that the intentionality of all types of constructs, can be broken down to the seeking toward the maintenance of dynamic stability which, for an aware human, amounts to maintaining stable dynamic conceptual interpretations regarding the phenomenon of qualitative conscious experience; whilst for a replicating organism, amounts to maintaining stable and therefore qualitatively relevant physiological adaptations.
B. I also argue that the various types of construct are all hierarchically related, in so far as their structural forms evolve in complexity, and that this evolution of complexity of form, ultimately leads to the transcendent emergence of its subsequent construct-class.
C. In some respects, one might say that my Hierarchical Construct Theory (HCT) reductively explains emergence. HCT achieves this by showing how each class of construct, firstly, obeys the same Newtonian principles and secondly, relates hierarchically.
D. Finally, HCT explains that any given structure pertaining to any particular construct class is informed about its environment and that this informedness constitutes a distinctive type of representation about its environment.