Mike Hosken's

"Hinges and Loops"



  1. Data Capture
  2. Data Errors
  3. Direct Sensing
  4. Data Transmission
  5. Data Processing
  6. Not a Lot of Sense
Right, then: down to business. The vast majority of ordinary knowledge - everything except instinct and revelation - gets into what we conveniently think of as our minds via our senses. As human beings we generally regard sight as the most reliable of the senses: we tend to accept anything that we've seen with our own eyes. And yet we know that we don't all see the world in the same way: some people have colour blindness to varying degrees; some need spectacles; and so on. It is similar with the other senses. Some of us are more or less deaf; when plagued with a cold we cannot smell; and so on. Modern technology tells us that there is a lot going on in the world - radio and television waves for a start - that our senses cannot register at all.

Acquiring knowledge by sight may seem to be achieved "directly". But, that is not really valid. It is worth taking a closer look (to coin a phrase).

§1.1 Data Capture

It is obvious that a Christmas card, say, doesn't get into the mind. In fact, it doesn't even get into our eyes: the most we can say is that an image gets into our eyes which our mind interprets as a Christmas card. So we are clearly dealing with a multi-stage process even with "direct" learning from sense data.
  • There is the physics required to form an image of the card on the sensitive surfaces at the backs of our eyes.
  • Then there is what we can nowadays think of as the on-board data processing and transmission as that image is changed to some form of message which is sent through to the brain.
  • We then have to use memory and experience to translate that set of messages into something that we recognise.
The number of opportunities for error is so great that the wonder is that we can rely on the system at all.

In diagrammatic form the shortest possible chain in my own case would seem to be something like this:

Christmas card

random light reflections in all directions

transmitted across some space and a moment of time

some light rays bent a bit by my spectacles

and bent again by my eye lenses

two slightly differing images on the backs of my eyes

patterns of reactions by the sensitive cells

message signals going along nerves

sorting-out of signals in and by the brain

combination into a single image

interpretation as a mental image

linking with parts of memory that may be relevant

comparison of mental image with stored data

"knowing" that it is rectangular
"knowing" that it has robins and snow
"knowing" that it is a Christmas card
"knowing" that it came from Bob and Mary

We could, I suppose, go on now at great length to examine every one of these stages. But that doesn't seem to be necessary for our present discussion. Be warned, though, that there are dangers in taking too much of it for granted. Many an earlier philosopher has ignored the dozen or so intermediate stages in that chain. He has tried to jump from the Christmas card itself directly to our internal knowledge of it. In extreme cases he denies the independent existence of the Christmas card, maintaining that it can exist only if/when/while perceived by some being or other: this is known as "idealism" - not, I may say, because it is ideal but because the idea of the card is regarded as the important thing rather than any card which may or may not actually exist. We shall not follow that road. Instead we shall have a look at the sources of possible error within the system.

§1.2 Data Errors

Most obvious, I suppose, is that in order to be able to see there must be some light: to see the colours accurately we need to be able to make use of a particular range of wavelengths since in physics colour equals wavelength. This is nothing like as simple as it may seem. In fact, of course, "physics" cannot see colours at all. Colour is something that you and I and many other living things can appreciate - though even we can't describe it. All that any physical instrument can do is to measure wavelengths and then it is up to us to match those figures with what we know, internally, about red, blue, yellow and so on.

Let's think about a favourite TV programme. Of course it doesn't really exist: it's all an optical illusion. It is quite well known that apparently smooth movement is actually achieved by showing a series of similar but slightly different still pictures in quick succession. The human eyes-brain system takes a minimum of something like one sixteenth of a second to realise what any particular picture is about. Some home movies worked reasonably well by projecting sixteen successive pictures per second: modern television uses half as many again in order to minimise flickering.

Not that your TV set can actually produce a picture: it can't! All it can do is to show one spot of picture at a time, with the spot location zipping across the screen one line at a time (taking something like one fifteen-thousandth of a second to do so), half-filling the screen from top to bottom and then going back to fill in the missing alternate lines before starting work on the next image. Meanwhile though, any watching human eye (I can't vouch for moths' or alligators') has been sending messages to the brain which "sees" a single picture - just one within the apparently smooth succession of whole pictures.

But if you enquire into the TV mechanism you quickly discover that it can boast of a mere three colours - red, green and blue. So can it "actually" produce mauve and gold and black and white and so on? Certainly a colour photograph of the screen shows the full range of colours so the answer would seem to be "Yes". But again it is just an illusion: size is responsible this time, rather than speed or time. The spots on the TV screen are of three types, each capable of glowing only red or only green or only blue. But as they are so tiny we can't make so fine a distinction unless we look at the screen very close-to, or with a magnifying glass perhaps. So on normal viewing the colour spots fuse and our eyes-brain system perceives the whole range of intermediate, mixed colours. Obviously, no colour at all equates with black: it is a happy fact of physics that equal contributions of the three basic colours gives white. (Incidentally, this is not the same as the colour-mixing system that you can achieve by mixing paints. In that case the primary colours are red, yellow and blue: no paint gives no colour and a mix of all three gives a dirty brown. The difference from the television system arises from the fact that the TV uses light production whereas paints rely on the reflection of light.)

There can be a whole host of other colour illusions. Generally speaking, our eyes-brain system compensates automatically for all but the most severe upsets of colour regimes: that's why your partner looks much the same in a dim restaurant, in sunshine, and in gaudy neon-lit surroundings. Researchers have shown that what seems to us to be a particular colour is not directly the result of the wavelength of its light alone: what matters is relative rather than absolute wavelength. If you want to demonstrate this for yourself you can try projecting your holiday slides onto a bright pink screen: the people and scenes appear to have quite normal colouring. But there are limits: don't expect your partner to look healthy and rosy under orange street lamps at night.

Even if we are seeing under ideal lighting conditions we can't say that we see "true" colours since the only colours we know about are human colours: colours are different for a bee since it can see (if we are still going to call it "seeing") up into ultra-violet frequencies beyond the scope of human eyes. And some snakes, for instance, extend their vision down to the infra-red "colours". So what are the "true" colours of the Christmas card? We can never know.

Similarly, it's as well to be aware of the cleverness of our brains as regards shapes: artists have to. Let us make a distinction:
  • How things seem. The actual images on the backs of our eyes or in photographs or in computer simulations are the result of perspective. The card itself, or a book say, virtually never seems rectangular because we are viewing it at an angle: the artist has to show items as they (actually) seem, with books squashed into parallelogram shape, and oval-cylinder beer cans.
  • How things seem to seem. Our brains re-interpret a scene or a photograph or even a drawing back to "true" shapes. Not realising this, a child doing a drawing has round things shown by making circular shapes on the paper: rectangular objects are shown using oblongs without perspective and consequently "looking" wrong.
The artist may deliberately use our brain skills to produce a bit of illusory nonsense.

Devil's Fork!

Even the direction of the light can make a huge difference to what reaches our eyes: highlights, reflections and excessive contrasts can make a thing "unseeable", as when the sun is in our eyes. There are not likely to be any detectable errors in viewing a Christmas card across the reading distance involved, or even across the width of the room. But transmission "errors" in other circumstances are many and very varied. Here are just a few to be going on with.
  • Looking along a hot road on a still day seems to reveal a wet patch but we know that it is really a mini-mirage, bending the light rays in the same way as happens for a full-blown oasis-in-the-desert mirage through heated layers of air.
  • Light from distant stars takes millions of years to reach us. So we cannot see how the stars are but only how they were when the light set off on its journey away from them, all that time ago.
  • When anything is moving towards or away from us we get the wrong information. This is most obviously so with the car horn or ambulance siren: the note drops as it passes us. Physicists call it the Doppler effect. This is not an illusion inasmuch as any appropriate scientific instrument parked alongside us would also report that the note actually does drop as the vehicle passes. But it is illusory in the sense that the horn or siren gives out a constant note: anyone riding on the vehicle hears it or measures it as remaining unchanged throughout. The same thing happens with light on the grander scale, so that most starlight is the "wrong" colour by the time it reaches us from distant parts of the expanding universe: it is redder than it was when it set off from the star - except of course for the light that started off red anyway, which is more invisible (to our eyes) than it should be.
  • The difference between light and sound gives rise to the "error" of hearing a distant gun fire long after the puff of smoke appeared. Or on a more modest scale, we might see that the batsman has been caught in the slips before we hear the sound of his stroke.
Of course those of us who do, wear our spectacles to "correct" our eyesight. It would be rather silly to say that we are deliberately falsifying our sense data. Perhaps we couldn't even read our philosophy books without their aid. At all events, we don't normally think of commonplace visual aids like reading lights and magnifiers as having any deep philosophical significance.

The case may not be quite so straightforward with other instruments though. Even a simple microscope shows us a world very different from that of our unaided senses. So we have to ask what the world is "really" like, rather than assume that our human viewpoint is in some meaningful way more true than is that of the carpet mite.

§1.3 Direct Sensing

Let us take it, however, that the Christmas card in question has reached the stage of forming two images on the backs of our two eyes. We've finished with physics: from here on in it's a matter of animal physiology - of how our body mechanisms function.

Although we tend to think of sight as being so much more reliable than the other senses it is perhaps worth realising that taste suffers none of the limitations we have been thinking about so far. When we taste something it is the actual material itself that has to be in direct touch with the sensors on our tongues. Even if some part of what we think of as taste is really the smell of the material the same rule holds good: minute amounts of the substance must reach the similar sensors in our noses. Only then can we form anything that could be regarded as a taste "image" of roast beef or strawberry yoghurt.

And yet we know for certain that people differ in the way they taste the same things. Geneticists are the scientists who study biological inheritance. For some purposes they are particularly interested in characteristics that are as simple as possible, controlled by a single gene, a single unit of inheritance. That makes it relatively easy to follow what happens to such a gene, both from generation to generation and in some cases from race to race or population group to population group. They happened to discover that the ability to taste a chemical compound called phenylthiocarbamide is controlled by a single gene: it is more common in some human populations than others. "Tasters" find it dreadfully bitter, while it has no effect on "non-tasters". The point is, from our point of view, that even in the case of tasting, where there is direct contact between what is to be sensed - the chemical in this case - and the sense organ concerned - the taste buds - the result is still not a reliably constant image, person to person. There could be some similar explanation as to why some people enjoy eating several green vegetables of the cabbage family and other people don't like any of them. Perhaps we experience the same thing in different ways: we can never know, however much we try to compare notes.

Smell is probably our least developed sense. Undoubtedly there is a lot going on in the world of smells that we don't know anything about. Any pet dog obviously has a picture of the world built up largely of smells - of olfactory information if you prefer. Certainly the bloodhound and the skunk, the dogfish and the salmon, the fly and the moth all rely on the sense of smell - their own or somebody else's - for finding food or a mate; for tracing the way home; for marking territory; or for defence. The concentrations of the chemicals involved are in most cases much, much too low for humans to be able to detect unaided.

§1.4 Data Transmission

But let's get on. Our discussion has reached the stage where something or other is affecting some sense organ of the creature in question. The next stage in the sensing process is all internal, concerned with messages from one part of the body to another. Fortunately we don't need to go into a tremendous amount of detail: all we have to worry about is whether or not these mechanisms of physiology are reliable.

And of course we know they are not! Well, not totally anyway. The classic case always quoted in this context is the poor chap who has had his leg amputated and yet it still itches from time to time: so he scratches his wooden leg! In fact, of course, the feeling which the brain has interpreted as an itch must have originated somewhere along the transmission path which used to come from the leg.

The mystery surrounding the relationship between physico-chemical signals in nerves and the mental processes that we attribute to our minds is a complex one with no universal agreement. We read of hallucinogenic drugs, of magic mushrooms, of alcoholics afflicted with delirium tremens. So we know for a fact that in common with all living processes the body's message-carrying system can be poisoned and upset in many ways.

Descartes was, you may remember, concerned about dreams. Certainly dreams and nightmares do not come from sense data in the ordinary way.

On all these grounds, then, we have got to realise that not every sensation was set up by a sense organ. We already know that sense organs differ in their sensitivity but once a sense organ has detected something does it necessarily send off its message and does that message always finish up as a sensation? Both answers are "No". We don't need to get terribly involved here: it gives us a good clue if we simply ask about sleep and attention.

One of the convenient marvels of our bodies lies in the matter of attention. Perhaps the most obvious example is breathing. We are "on automatic" virtually all the time, so we don't have to give any attention to it: the control system keeps us breathing fast enough to provide all our body systems with the oxygen they need. But the moment we want to drink, swim, kiss, gargle, or blow up a balloon - in fact whenever we need conscious control of breathing - we can switch to "voluntary" for as long as we want.

The same thing applies to the incoming signals from our sense organs. If you stop to think about it - to give your attention to it - here and now, you will realise that many areas of your skin can feel your clothes; you know pretty clearly where your arms and legs are; you would know immediately if the bottom of this page suddenly turned red, and in fact your eyes have images of much more than the few words you happen to be concentrating on in the normal course of your reading; there may or may not be noises in the background, but certainly you would be likely to be aware of it if your ears changed their messages to your brain; and it could even be that you can, when you think about it but not otherwise, still taste your last meal and can smell something in your environment. The plain fact of the matter is that at any time, even when we are asleep, there are countless sensory signals being received and being processed which never reach any level of awareness. This is just as well: I would guess that we would suffer nervous exhaustion after a couple of days' attention to everything!

To make our living simpler (but our understanding of the process more difficult) there is no straightforward link between the strength of the sensory data and the level of attention. I'm not here referring to blindness, deafness and so on: quite apart from that there is no direct relationship between the strength of the signal reaching the brain and the brain's reaction. It seems as if receptivity can be pre-programmed. The fortunate and well known result is that the conscientious Mum can hear and distinguish her baby's cries when for anyone else they would be lost in the background noise or the soundtrack of the television film. Any of us can "keep an eye open" for some object or "be on tenterhooks" for a significant signal. Conversely, it is very difficult indeed to guarantee that a particular sound did not take place, that a special person was not among those who have walked by. Even worse, it is virtually impossible actually to detect insignificant events: would you know whether or not an aircraft had flown overhead while you were changing the oil in the car or planting your new shrubs in the garden? Normally, of course, it doesn't make the slightest difference: after all, we are discussing insignificant events. The worry arises only if you are witness to some crime, say: then your brain is not properly attention-programmed and it is very difficult to swear what did and what did not happen.

Anyway, enough about attention. What we must be aware of is that sense organs are not like parking meters, clicking reliably into their repertoire of reactions at distinct pre-set levels.

§1.5 Data Processing

All things (so far) considered, the wonder is that we are able to trust our senses at all. But let us get back to the case of the Christmas card, assuming that we have got to the stage of producing some kind of mental image. How we then digest, analyse or file that image depends on another whole host of things. Like, for example, which bits can be recognised? This shouldn't present too much of a difficulty with our Christmas card: we can reasonably expect to have amongst our mental equipment enough matchable units to be able to read the letters and assemble them into words and greetings; to recognise a robin whether shown true to life or as a symbol; and to link the "Bob and Mary" phrase with our memories of a couple of people we met during the summer. But it is as well to realise that all this depends on memory - memory of things like robins, memory of personalities like Bob and Mary, memory of letters and memory of words.

If a report of prices on the Tokyo stock exchange arrived into our environment in the form of an explanation of the Nikkei Average in Japanese few of us in the west would be able to get the meaning out of the given data: the image-to-meaning conversion could not be carried out. But the case would be exactly the same if the Christmas card artist had gone a bit over the top and made the symbolic form of a robin rather too way-out, remote from its real-life shape and colour: we couldn't then recognise that either. Certainly the interpretation and recognition parts of the system are very obviously significant when we contrast our own "seeing" with that of an expert. What is just a patterned stone to me is an identifiable fossil with an imposing Latin name to a palaeontologist. And my dentist can learn a lot more about me than I can from a view of my dental X-ray pictures.

We all accept that an ordinary landscape photo is an accurate image of a scene: it doesn't need any "interpreting" - it is just as it is. But then you might come across the interesting case of stereoscopic pairs of photos. If these are carefully lined up in a special mount with eyepieces (a stereoscopic viewer) it is possible to "see" in three dimensions. Of course the photos are both still flat: the 3-D effect is produced by making use of the brain's powers of interpretation. It's not the actual eyepieces of the viewer that make the difference but the alignment: the brain can't fuse two images which aren't at least partly identical. (The 3-D view is also lost if ordinary field binoculars are badly adjusted, but in that case the scene is in 3-D whereas the flat photos could show depth only as an illusion.) The photos don't change when they are aligned but the brain's use of them does. Incidentally the 3-D shapes can be "wrong" if the viewer eyepieces are not the same distance apart as were the lenses of the double camera that took the photos. So the brain might be illusorally right or illusorally wrong!

In short, a lot of information can be present and available, correctly imaged by the sense organs, sending valid messages to the brain, without correctly yielding its data unless handled by the brain in a particular way or in a special context of interest and attention.

§1.6 Not A Lot Of Sense

The sense data that we actually use to control our lives seem not to be up to the job of providing us with reliable knowledge. But apart from genetics and revelation we have no other way of knowing and learning things. My senses are my ONLY link between the real internal me and the world about me: they are my one and only me/environment interface.

If sense data by themselves are not enough to provide true knowledge there is a pretty clear need for something else, either instead of sense data or in addition. This cannot be anything external to ourselves since the only route back to ourselves is via our senses. But we have got to take account of the existence of the world and must accept something of the knowledge we get from our everyday senses. So we have got to conclude that the only possible way of improving the status of our supposed knowledge must come about from organising the sense data in some way. Put another way -
  • We must take account of the world around us.
  • The only way of learning about the world around us is by making use of our sense data.
  • Our sense data are unreliable - incomplete, illogically selected, liable to inaccuracy and subject to misinterpretation.
  • THEREFORE we must do what we can to improve the situation.
    • It could help if checking and editing were conscious.
    • The resulting knowledge must be self-consistent - as far as possible.
Some philosophers in their ivory towers have tried to deduce what the nature of the world must be by pure intellectual logic: they turned out to be wrong. We can't do without our sense data: we shall have to find some way of screening or reinforcing them if we are going to finish up with any greater faith in their relevance to true knowledge.

< < < < < < < < < <+> > > > > > > > > >

Please contact Mike Hosken at
[email protected]
with your observations, comments, criticisms and suggestions, or to request an A5 printed copy of "Hinges and Loops".
The next chapter concerns Number Puzzles.
Hinges and Loops CONTENTS