`If consciousness is why, intelligence is how.' Let me illustrate with certain differences I have noticed between cats and dogs. Just a few moments ago I watched my cat jump up onto my wheelie bin, about a meter high, then from there to the top of a two meter-high door. This door is only around 3 cm thick, and it swung slightly as the cat landed on it. Far from losing its balance however, my cat then walked backwards and forwards along it twice, then actually sat still on it for about three minutes as it peered around in the way cats do, whiskers aquiver. For reasons known only to itself, it then jumped back down to the floor again via the bin, curled itself up in a plastic box not much bigger than itself, and dozed off.

I can really only speculate here, but was this the last vestige of its originally survival-enhancing ability to jump from a rock directly onto the branch of a tree to pursue a prey animal, presumably a bird which even my modern-day cat can still catch? Like most other cats of my acquaintance, it also has a predilection for chasing the end of a piece of string pulled jerkily along the ground. Perhaps the cat perceives this as behaving like a mouse or other such small prey, just as one or two quick lines on a page can evoke for us the image of a mouse.

A dog chasing a ball does not appear to be anywhere near as clever, indeed to us it seems almost a dumb thing to do, even if charmingly so. But a few years ago I watched a dog do exactly this dumb kind of thing in a way that can only be described as intelligent. It was running after a ball that had gone over a low wall, about half a meter high. But on the other side of the wall, invisible to the dog, was a two-meter drop to the beach below. Three meters to the right of the dog, the wall had a gap in it. From this a flight of steps, also invisible to the dog, led down to the beach.

The dog first looked at the wall where the ball had gone over, then at the gap leading to the steps, then at the wall again. It repeated this behavior two or three times before finally deciding that heading for the steps was the `best' choice. But how could it have worked this out, since it could see neither the sheer drop on the other side of the wall, nor the steps? I can only assume that the dog lived locally and had a `map' of the area in its head, including all likely hazards, and could make its decision using this map. Certainly one could say that dogs, along with many other species including rats (and possibly cats) evolved such a map-making capability many millennia ago and that it's `decision' was based on this. But considering the time it took, the process of making it was clearly complex and it looked to me like it was `thinking'. Presumably it weighed up the risk of injury (it may have been an elderly dog, I don't recall) involved in the quicker direct route, versus the longer but safer route to its `prey'.

In short, the pursuit of prey was the `why' and determines the nature of the consciousness of both animals, the `weighing up' and the subsequent decision was the `how' and defines the dog's intelligence at least. What might a non-local dog have done? Would a cat be capable of making a similar decision? (I have seen a fat elderly cat demur at jumping off a verandah rather less than a meter high.) And while cats may be more agile than dogs, there are certainly things they can't do. Mine for instance can't apparently recognize itself in a mirror, even as another cat, otherwise it would surely respond in some way, perhaps as if threatened.

In the human world, the differences between (albeit somewhat nascent) consciousness and intelligence are similar to those between goal-seeking robots (like those that take part in `Robotics Contests'), and chess playing or `artificial expert' programs. The robots have sensors and possess the ability to react via their effectors to what they are programmed to do with their sensory data. The fact that the robots and their programs are designed and built by humans is almost beside the point, `natural' systems operate the same way even though are honed via `designed and built' via natural selection (as you can see with insects at least). It is our robots that seem most likely to eventually acquire consciousness, which we will perceive as a propensity to seek goals of their own rather than those we program into them.

Systems that play chess or behave like expert systems on the other hand simply mechanize what we do ourselves like any other machine; they extend our intellectual behaviors rather than our physical ones. Indeed here we might do well to distinguish between `intelligence' such as we saw displayed by the dog, and `intellect', generally a measure of how efficiently an individual `survives' within our culture (perhaps what IQ tests and suchlike really measure). Or to put it another way, intellect enables us to apply algorithms, intelligence (in combination with intellect) enables us to create them.

But before cats, dogs, humans, and the conscious machines of the perhaps not too distant future can apply their intelligence, they must be able to translate the `raw data' from their senses into something they can act upon. The first part of PereGaea (1: PereGaea, and I won't hyperlink these chapter references, the book really does need to be read sequentially) deals with just this question. It's `robots' use a vision system discredited for its unreliability in professional AI many decades ago (2: Vision). But PereGaea is different from an AI lab, here (as on Earth) it doesn't matter if perceptual (and response) systems are crude and error-prone provided sufficiently large numbers of organisms survive long enough to reproduce and spawn more such numbers. As sensors become more efficient (3: Lists, 4: Maps) fewer such organisms need to be born with each generation, and once they acquire the ability to learn from experience (8: Decisions) and plan ahead (9: Tactics), tens rather than hundreds or thousands of offspring with each birth may suffice. The ability to make decisions also marks the point where mere sense-response automatons acquire the first vestiges of what we would observe as consciousness; the tactical planning ahead that of intelligence. Mutual assistance through the evolution of communication between individuals (7: Signals) provides a wholesale leap towards collective behaviors (10: Culture) and the technology that may then develop (11: Artifacts). On PereGaea at least it is only at this point that language and intellection can then develop (12: Words). Signals, which many species including insects use (via sound, pheromones, bodily posture), ought not be confused with language, which is almost exclusively the human domain. For us it is the difference between the words we use, and the tone of voice and facial expressions with which we accompany them (and how they can quite often conflict, something we might fail to observe at our peril). Our emotions are our signals.

But of course before we can use words at all, we must define them. And that doesn't just mean words like `consciousness' and `intelligence' but those associated with them such as `sentience', `object', `recognition', association', `reaction' et alia. PereGaea sets out to do exactly that as best it can be done. But it doesn't use just words, it also employs a means of communication which, as it suggests itself, may well be far more ancient: drawings, around four hundred of them. Perhaps with the advent of cheap computers, drawings will once more become as valid an instrument of intellection as they were in those far-off times.