Imagining Itself (Part XV: Capability and Knowledge)

Can we become capable of doing things that we are currently unable to do, simply by thinking of them, by imagining ourselves doing them? Is imagination an enabler of action? Could I become capable of making a violin simply by carefully imagining the whole intricate process? The answer to these questions should be obvious but the underlying reasoning will require a certain amount of patient chiseling and shaping to be carried out beforehand.

Ten years ago I decided to make a kitchen table. I knew I had all of the basic skills necessary to start the job and I also knew that with a little research and care I could acquire the further skills necessary to overcome any of the foreseeable obstacles I might encounter. In short, I was certain that I could make a simple table and this gave me the confidence to be a little daring and attempt to learn on the job. The process was slow and I made many mistakes (fortunately none that I couldn’t fix or replace) but eventually through many unexpected twists and turns I completed what I still consider to be a handsome piece of oak furniture that is in daily use. Buoyed up by this success I decided to embark on the more ambitious venture of making a double bed from cherry wood. Once again I made several silly but salvageable mistakes along the way but eventually ended up with a simple but elegant piece of furniture held together by 52 hand cut mortise and tenon joints. It’s a thing of pride and an object I am obviously intimately acquainted with.

Neither of these ‘projects’ would have been possible if I hadn’t already acquired the skills (and tools) to at least commence them. However, even without those skills I could easily have imagined what it would be like to make these objects. But then, imagining is not knowing - and knowing, as we will soon find out, is not necessarily the capability many people are inclined to think it is.

Around the same time as I was chiselling mortise and tenon joints, Derek Melser, a furniture-maker-turned-philosopher living in New Zealand, published a book entitled “The Act of Thinking”. The underlying thesis is very similar to that outlined in Part XII of this series of blog posts, and to that extent I think he has got it absolutely right: thinking is a species of action. Nonetheless, there are several respects in which Melser’s theory doesn't adequately explain the less physical of our actions – most especially he is curiously vague on the subject of how imagined perceptions might constitute actions. For example, what might an ‘actional’ visualisation consist of? Melser writes: “To ‘visualise’ thing T is to covertly token a certain visual perceptual behaviour.” To “token”, for Melser, is to enact only a fragment of an action and in doing so the token becomes a referent to the thing tokened.

The argument that some inhibited tracking movements of the eyes, or some tokened verbal descriptions of the things or experiences imagined are sufficient to explain visualisation is unconvincing. So far as gesturing and speaking are concerned I think Melser may be largely, if not wholly, right (I’ll return to this presently). But as to his account of imagined perceptions, I think a more expansive explanation is due.

To be fair to Melser, he stages the different performances that comprise his overall thesis with genuine skill and he directs the various actors expertly, making them speak to each other and to us with close attention to nuanced argument and overall coherence (which is to care for the audience with clarity – a commendable thing in any philosopher). Where his workbench is a little shaky though is in its incorporation and understanding of representation and representational strategies.

Melser takes the view that perception is achieved “when and only when, […] an appropriate verbal act is performed” such as the infant’s exclamation of “mummy!” at the appearance of her mother. Melser is partly correct, I think, but he makes the mistake of overemphasising verbal representation  (or “concerting” as he calls it) at the expense of other equally valid forms of representation.

My son, who will be 3 very soon, is still learning to name colours. Does this mean that he doesn’t perceive them? On Melser’s account we have no option but to conclude that he doesn’t yet perceive colours but I can prove this is incorrect with a simple experiment that I tried more than 4 months ago. I set up 5 different coloured objects and asked him to find others of the same colour. He got it right every time. We have already encountered an answer to what is happening here provided by Donald Brook’s theory of representation. Children are able to select Matching and Simulating representations long before they are able to speak their names.

I agree with Melser that the ability to represent something is a precondition for perception. But the skill of representation is by no means first acquired through our entry into language. If a forthcoming paper by Donald Brook is anything to go by, the capacity to represent, in rudimentary form at least, is also shared by many animals and possibly some insects also (bees for instance) which suggests that there must be a genetically inherited component at work. Furthermore, if thinking is a form of covert action, then who knows how many animals might be capable of rudimentary forms of representation? This is a question only science can answer.

Melser writes: “One of the main features of imagining is that you can do it where real X-ing [seeing a ghost for example] is impossible.” This seems perfectly right doesn’t it? I can imagine jumping to the moon but I can’t do it. But if imagining is a species of action then what imaginary action could we possibly ‘do’ to visualise a ghost? The problem is one that Melser’s theory simply cannot solve. However, if we expand the conception of action to include representational action then suddenly the whole difficulty evaporates. To Imagine a ghost is to imagine what a ghost would look like i.e. how we would represent a ghost to others, for example by cutting holes in a sheet or doodling a white image on black paper or by wafting steam about etc.

So, to return to the question posed at the beginning of these bloggy thoughts: Can a capability emerge as a consequence of imagining?

If imagining something is a process of representationally oriented action, then to be capable of representing an ability is no guarantee of the capability of doing it. The capabilities of Matching representation that involve bodily motion and control (i.e. gestures, postures, facial expressions etc.) on the other hand are genuine proof of ability. If I can mimic your dance steps, footfall for footfall, then there is no question that I know how to perform your dance. But if I can describe your dance, footfall for footfall, no matter the intricacy of the detail, there is no guarantee whatsoever that I can follow my description. Different forms of representational ability presuppose capacities, but most commonly these are capacities of representational action, not of performance.

Imagination is a form of what Ryle would call “knowing how”. Too often people confuse the knowing how to represent with the more practical capabilities of knowing how to do.

“A child who had never manifested in words, gestures, or play the working out of simple problems could not be said to work them out ‘in his mind’, any more than he could be said to know ‘in his mind’ the names of colours, if he was unable to say their names, or to point or to fetch the right colours when their names were called out. Thinking in ones mind (silent thinking, pausing to think) is not the most fundamental form of thinking, but instead presupposes thinking in play, work, or words.” -Norman Malcolm

Imagining Itself (Part XIV: Visualisation Disability)

Is it possible to have an imaginative disability; an impairment or deficit in or of the mind’s eye?

Before she died and for somewhere in the region of a quarter of her life my grandmother was blind in one eye (one of the physical one’s that is, not the 'inner' one). She could easily have arranged for a second cataract operation to restore her stereoscopic vision, yet she was too distrustful of the doctors and too unconvinced of the benefits that she chose instead to manage with only monocular vision. I remember once at the family dinner table we all decided to compare our monocular skills by trying to replace the lid of a pen and my grandmother was by far the most skilful at this task. She had evidently developed compensatory skills that allowed her to function very well despite her depthless vision. For many people such a disability would be a great loss - the neurologist Oliver Sacks has commented widely on what he believes is a great impoverishment in his visual perception following a melanoma that deprived him of vision in one eye - but for my grandmother the possibility of renewed depth perception wasn’t even worth a short visit to her local hospital.

Disabilities clearly affect people in different ways and nobody would seriously suggest that all disabilities are equal. Having only one functioning eye is undoubtedly a disability but in comparison with complete blindness it presents a relatively trivial setback. But what about loss or damage to the “third eye”? What might such an impairment or deficit consist of and does it really make sense to call it a disability at all?

In 2010 Discover magazine published an article about an Edinburgh building surveyor MX who found that his inner eye had suddenly become blind following an operation on his coronary artery at the age of 65. Neurologists conducted a variety of experiments and scans to determine what might be causing MX’s visualisation loss and they, and MX himself, were surprised to find that only one of the many experiments they conducted showed any marked difference from standard results obtained from other individuals of the same age, profession etc. The test used is thought to require the ability to mentally rotate diagrams of three-dimensional forms. Ordinarily test subjects take twice as long to mentally rotate the diagrams 180% as they would to mentally rotate them 90%. In MX’s case though it took him no longer to rotate a diagram from 180% than than 90% yet in every other respect there was no appreciable divergence between his visual performance and those of normal individuals.

How might we interpret this account? The scientists involved took the view that some form of blindsight must be involved. Blindsight is a rare condition in which people have no conscious vision yet they are able to perform certain tasks – even quite complex ones like walking through a room full of obstacles – as if they were fully sighted. A common explanation of blindsight  – although further evidence is needed - is that human visual perception involves two visual systems working in tandem: a more recently evolved conscious system and an older more primitive unconscious system that remains unaffected in cases of blindsight.

Just as with the difference between monocular vision and total blindness a straightforward comparison between blindsight and third eye blindness should probably be avoided. Blindsight is recognised as a disability because it has very clear repercussions for visual functioning whereas, if MX’s case is anything to go by, the impact of visualisation loss is comparatively negligible.

Nonetheless, perhaps we shouldn’t be too hasty in dismissing third eye blindness. According to the Discover article, MX felt his inner vision had previously been a source of genuine pleasure that allowed him to run “through recent events as if he were watching a movie. He could picture his family, his friends, and even characters in the books he read.” But this is where the account becomes a little unclear. Had MX become unable to recall the appearance of his friends and family and of things seen in the past? If he was suffering from some form of visual amnesia then surely the researchers would have detected this right from the outset. Other plausible explanations are possible though. Perhaps MX’s capacity to derive pleasure from visual cognition had been affected by the operation. Perhaps his sense of past pleasures had somehow become exaggerated or distorted.

One of the things that makes discussions about mental imagery so fraught is that people have very different emotional attachments to such aspects of their subjectivity. Just as Oliver Sacks deeply regrets the loss of his stereoscopic vision and my grandmother barely bothered about it, so too do opinions differ greatly about what might well be essentially the same underlying phenomenology of visualisation.

Certainly when individuals claim to ‘see’ vivid images in their mind’s eye we should be sceptical of what it is that they actually mean. Time and again when such claims are tested, the images reported turn out to provide much less information than their owners are initially willing to claim. Once again the difficulty would seem to derive from the feeling of vividness rather than the quality or quantity of information available.

Whatever the differences between individuals in terms of the accounts they are prepared to give of their subjectivity it would seem to be vital to distinguish as sharply as possible between the emotional aspects of visualisation and the function: the ability. If visualisation doesn’t provide any kind of functional advantage then it makes no sense to call it an ability, though we might very well call it a pleasure. Likewise, being unable to visualise cannot be said to be a dis-ability.

But if it is true that visualisation consistently fails to provide the abilities that are attributed to it then it must be unique amongst all pleasures. Perhaps this offers us a vital clue. Evolution never bestows pleasures unnecessarily, especially pleasures of the magnitude of imaginative visualisation. MX may have functioned perfectly well on all of the tests of visualisation ability but I wonder how much his motivation to create and consume representations had been affected – I suspect it was greatly reduced. So, perhaps we can put forward a speculative hypothesis: visualisation might not be an easily quantifiable ability in the ordinary sense but perhaps it's utility is of a more pervasive kind; as an inducement - possibly the most powerful one we possess - to make, describe and to consume visual representations.

Imagining Itself (part XIII: Intentionality and Intention)

It is of the very nature of consciousness to be intentional and a consciousness that ceases to be a consciousness of something would ipso facto cease to exist. -Jean-Paul Sartre,

Whilst examining the differences between photographs, caricatures and mental images, Sartre writes:

The material of the mental image is more difficult to determine. Can it exist outside the intention? […] In our opinion, it is not only the mental image which needs an intention to be constructed: an external object functioning as an image cannot exercise that function without an intention which interprets it as such.

Here Sartre is drawing upon the earlier work of influential German philosopher and psychologist Franz Brentano who held the view that all mental states exhibit an intentionality or ‘aboutness’ i.e. they are directed towards an object, albeit one that is “inexistent”. This same notion of intentionality has also been a popular reference point in theories of representation – though its popularity seems to have waned in recent decades  - and it is clear from the above quote that Sartre conceived of representations in the same way: as things that can only function through interpretation. As we have already seen a more thorough theorization of representational strategies provides a far more illuminating means to understand and examine representations.

So why bother with intentionality? The main reason for addressing the idea of intentionality in a discussion of imagination is because intentionality is widely regarded as fundamental to the phenomenon we call “mental imagery”. Nigel Thomas makes this very point in his detailed entry for the online Stanford Encyclopedia of Philosophy on the subject of Mental Imagery. There is certainly no need to hold back the tide of opinion on this matter but what we must do is establish a clear distinction between intentionality on the one hand and intention on the other. For the purposes of this discussion we will take an intention to be a mental state directed towards a goal as a necessary component in purposeful action. It will also be necessary to determine whether the notion of intentionality applies universally, not necessarily to mental states but to behavior, much of which we take to be the consequence of cognitive - though not necessarily conscious - processes.

For a 2011 book-chapter entitled “Evolutionary Emergence of Intentionality and Imagination” philosopher Dale Jacquette discusses intentionality as a cognitive activity directed towards (or ‘about’) an object:

For a living thing to sense or perceive is for it to sense or perceive something, to be directed in psychological occurrence towards the object intended by the sensation of perception.

Jacquette outlines an example of a mollusk on the seafloor snapping its shell closed in response to a change in light or motion. He writes:

The internal neuronal change triggered by environmental events will accordingly have an intended object, even in so simple an organism, as a proximate cause of the neuronal state change. The cause of this neuronal state change is in turn what the change intends or is about.

This can’t possibly be right. The mollusk doesn’t have an intended object, it is simply responding to a stimulus. The response is not ‘about’ anything, it is not directed ‘towards’ any thing, it merely moves. To impute an intention to this primitive motion is to take a giant leap into error. If intentionality is directed toward any ‘thing’ in this instance it is toward the result of a movement, not an object. If we were to follow the preposterous logic of Jacquette’s argument then every smoke alarm, not to mention every mobile phone, would have some intended object as its ultimate aim.

The mistake Jacquette makes here is precisely the same as that made by Alva Noë in his assumption (already discussed) that bacteria are on a behavioural spectrum with the complex behaviours of humans. If we are to meaningfully distinguish between the sometimes highly sophisticated response systems of basic organisms and the far more complex behaviours of higher forms of life then we will need to understand the role not of intentionality but of intention.

For a behaviour to be intentional there has to be an anticipated goal towards which the behaviour is directed. If no goal is involved we would simply have movement or an advantageous response or reaction of the kind commonly exhibited by mollusks. Mollusks do not choose between contemplated options. They do not form an alternative goal and respond differentially based upon this anticipated future state. Humans on the other hand do this a great deal. In fact we form intentions and entertain actions and refrain from enacting them on a massive scale. Take for example the familiar process commonly referred to as “inner speech”. To conduct an inner monologue or silent soliloquy is, I suggest, to enact absolutely everything aside from the physical performance of speech itself.

Imagine someone unable to ‘hear’ their inner voice, someone who can only consciously think by speaking everything aloud. Such people do in fact exist. Pre-school children commonly engage in ‘private speech’ and it is only around school age that this becomes internalized or, more accurately, the speech stops but all the other associated processes continue.*

So, strange as it may seem, consciousness is not where the majority of mind-work is being done. As is already well understood, a vast majority of decisions are made long before they ever find their way to conscious awareness. What we call consciousness is in fact just the process in which we privately enact (ie: inhibit from public performance) everything but the representations themselves. If this seems hard to believe, then it might be worth considering the conscious thought processes of the congenitally deaf. Congenitally deaf people brought up to sign do not report having inner voices, they report having inner signing: they ‘feel’ and ‘see’ themselves gesturing ‘in their head’. Obviously there are no physical gestures going on in their heads, nor are there representations. What is going through their minds are the intentions to represent via signing.

Conscious thought then, understood from the perspective of intention, might be best characterized as a form of covert (i.e. unexpressed) representationally oriented action. Once again, this is not to side with those who claim that we experience inner representations. This is exactly the suggestion I have been trying assiduously to avoid throughout these discussions on imagination. The claim I am making is as follows. We are a social species and as such we have evolved a range of representational practices that allow us to communicate with one another in highly sophisticated ways. The importance of communication is so fundamental to us and the associated dispositions to represent are so thoroughly embedded that we can barely distinguish between the intention to represent something and the publicly perceptible representations we are capable of producing.

We think we imagine in words, images and propositions etc. when in fact we imagine in intentions: intentions to represent the thing or event in question. So, to speak out loud (i.e. to make verbal representations) and then to continue this speaking as an inner monologue is not to continue the representations in one’s head. It is to terminate the speaking but to continue with the intention: i.e. all of the mental actions apart from representation itself.

We might say that imagining oneself talking or humming is a series of absentions from producing the noises which would be the due words or notes to produce, if one were talking or humming aloud.” –Gilbert Ryle

*It might also be interesting to note that the practice of silent reading is also a relatively recent phenomenon. In “A History of Reading” (1996) Alberto Manguel unearths several rare descriptions of people reading silently but he points out that: “not until the tenth century does this manner of reading become usual in the West.”

Imagining Itself (Part XII: Sartre’s Imagination)

In 1940 French philosopher Jean-Paul Sartre published “The Imaginary: A Phenomenological Psychology of the Imagination” in which he expounds his theory of imagination. Sartre is frequently cited as a prominent critic of mental imagery - as indeed he was - but his work is nonetheless far from rejecting mental representation altogether. For Sartre what we call a “mental image” is in fact a jumble of interlaced and simultaneously available representations, or what he calls an "Analogue" of perceptual experience. According to Sartre, when we imagine something our consciousness is not directed toward an image but rather the object is directly apprehended:

The imaginative consciousness I have of Peter is not a consciousness of the image of Peter: Peter is directly reached; my attention is not directed on an image, but on an object.

In neither case (perception or imagination) is the object – a chair for instance - actually physically present in consciousness, that would be impossible. Instead we have "a certain type of consciousness, a synthetic organisation, which has a direct relation to the existing chair and whose essence consists precisely of being related in this or that manner to the existing chair." He argues that perception and imagination are identical relations between consciousness and reality.

Despite his criticisms of mental imagery, it is strange that Sartre should be so willing to use the term "imagery" extensively to refer to these figments throughout his book. Mary Warnock, in her 1972 introduction to "the Imaginary", also raises issues in this respect:

…so what is this object of my consciousness? To what am I attending? We have to remember that Sartre has said that there can be no image in the mind. But it is here that he genuinely appears to vacillate.

To be fair to Sartre, it is not in the least surprising that he struggled over the question of mental imagery. It is a subject that has vexed numerous eminent philosophers and scientists alike and continues to do so. So, rather than dwelling on this weakness in his theory, perhaps it would be more illuminating to ignore his vacillations and focus our attention elsewhere.

One of Sartre’s more radical claims is that we can learn nothing from mental imagery. In comparing mental images with the “overflowing of the world of things” – of perception –  he writes:

“The image [i.e. mental image] teaches nothing: it is organised exactly like the objects which do produce knowledge, but it is complete at the very moment of its appearance. If I amuse myself by turning over in my mind the image of the cube, if I pretend that I see it’s different sides, I shall be no further ahead at the close of the process than I was at the beginning: I have learned nothing. […] No matter how long I may look at an image, I shall never find anything in it but what I put there.”

Warnock finds this “neither entirely clear, nor, as far as it is intelligible, strictly true". She gives the example of envisaging some previous acquaintance to establish whether or not he has a moustache and she finds that this “example suggests that we may sometimes believe ourselves to be able to find out more about something from our image”. Warnock may be right that Sartre is not entirely clear on this issue but I think something important may be falling between the cracks. No doubt Sartre would agree that it is sometimes worthwhile striving to recall what we barely remember, and trying to ‘picture’ – as we say – whether someone does or doesn’t have a moustache may be an excellent way to tease out a residual memory. But the point Sartre is making is a more profound one I think. He is engaged in distinguishing between perception and what he occasionally - and helpfully - terms the “quasi-observations” of imagination. For Sartre what we call mental images are the products of memory – that is to say – what we have experienced and remembered, and in this sense they are unlikely to teach us anything we do not already know. If imagination is fuelled by memory, then it would seem to be perfectly consistent to say that we “shall never find anything in it but what [we] put there”.

But where Sartre oversteps the mark, and where Warnock is right to question his claims, is on the subject of our ability to learn from our imaginings. If to imagine were simply to remember something then he would be correct and there would be nothing further to be discovered. But imagination is not simply a process of linear memory recall of uninterrupted episodes. If it were so, then why distinguish between memory and imagination? Unlike straightforward memory recall, imagination allows us to recombine and compare fragments of memories in order to form inferences, to plan, anticipate, and problem solve. If through such cognitive recombination we arrive at realisations that had previously been unavailable to us, then I think it would be true to say that we do indeed learn, or at the very least we come to form ideas and intentions that have the potential to direct our actions in ways that lead to productive learning.

So, we should probably be a little skeptical of Sartre’s claims. He undoubtedly has a lot to offer in terms of raising and discussing some important issues at a time when few others were interested. But despite its seeming radicalism, Sartre’s view turns out to be somewhat less illuminating than might be hoped. Having said this, he does have one or two important things to say about the relationships between intention and imagination. This will be the next stop on our journey through the neural networks of the imagination and as we begin to look a little deeper into the subject we should see how intention plays a pivotal and widely recognised yet poorly understood role not only in imagination but in consciousness also.