Symbolic reasoning demands precision: Symbols can come in a host of different orders, and the difference between (3-2)-1 and 3-(2-1) is important, so performing the right rules in the right order is essential. Marcus contends this kind of reasoning is at the heart of cognition, essential for providing the underlying grammatical logic to language and the basic operations underlying mathematics. More broadly, he holds this extends into our more basic abilities, where there is an underlying symbolic logic behind causal reasoning and re-identifying the same object over time.
When I read stuff like that I find it wearying. It requires intense application on my part to figure out exactly what is being said. My summary of it, after a brief and no doubt glib analysis, is this: the mechanism (grammar) for insight (cognition) is basic (underlying) logic (axioms and repeated confirmation). This means we learn by what is self-evident and by what is recalled. Intelligence is thus casual (off-the-cuff) as implied by the observation, “What was your first clue?” Naturally this level of Artificial Intelligence (AI) is appropriate to basic intellectual abilities distinguished from studied knowledge (learning by definition) such as mathematics, Latin or any other language, economic theories and legal maxims. Law is certainly dedicated to prescriptive thinking; that is, governance by statute, precedent and deductive reasoning generally. That type of understanding is almost scientific in that the rationality is axiomatic, at the very least predictable.
Deductive reasoning, the mental process of drawing inferences in which the truth of their premises ensures the truth of their conclusion (if A = B, and B = C, then A = C)
The arena of AI which attracts the greatest attention is therefore the thought arising from patent and acquired knowledge. The algorithms of a computer can be trained to adopt this method of thinking in the same way we do naturally. Interestingly however the methodology embraces what is simple logic not some complicated scheme. It is a process which echoes the clarity one recognizes with age and so-called refinement.
Training a computer not to make an inductive leap is no easier than training one’s own mind to do the same. It is at least a reminder that we mustn’t see or think what we prefer in lieu of what is happening. Again, another simple rule, “Believe what you see!”
Simplification is in my opinion at the heart of the issue. Nature, as marvellous as it is – especially at this time of year Springtime – cannot be construed as so complicated to obstruct so- called “casual reasoning”. I mean, if a dandelion can do it, there’s a good chance we can interpret its “underlying symbolic language”.
The brilliance of simplicity is not merely its lucidity; it is its pervasiveness. This in turn implies that the elemental truths of the universe are in many respects common and shared, that they exemplify identical theses, recognizable methods of precision. There are I am certain many features of nature which reappear throughout its myriad of expressions.
Spotting these elemental features – “the same object over time” – and conjoining them with “underlying symbolic logic” is the key to enlightenment (“cognition”). The machinery is the grammar of language or the rules of mathematics which respectively enable growth (variance) but demand precision (clarity). If we pay attention to these fundamentals – even if promoted artificially – they will (almost by definition) afford erudition and maybe even wisdom. I believe we’re missing the boat if we presume that our “more basic abilities” are somehow distracted from or superior to this mechanical description of intelligence. The rules of the road are the same wherever one goes. The hurdle if any is that there are a lot of different languages to express what we pinpoint; what preserves their clarity is their simplicity.