[Adaptive Thinking: Rationality in the Real World, Gerd Gigerenzer, Oxford University Press, 2000]
Professor Gigerenzer starts off the introduction to Part IV with a strong claim:
The study of human thinking is deeply suspicious of introducing anything genuinely social into the world of “pure” rationality. As in much of cognitive science, most researchers have fallen in love with syllogisms, first-order logic, probability theory, and other systems that abstract and distract from the semantics and pragmatics of thinking, not to mention social motives. [page 199]
Before passing on, I’ll ask the reader to ponder this question: if thinking has no social aspects, then what has rationality to do with morality? In fact, we modern human beings are in the embarrassing position of having our major philosophers claim that virtue, vice, and even Christian views of sin are metaphysically founded while most of our public behavior is socially and biologically motivated. Virtue must be imposed upon creatures of flesh where the flesh and the social relationships of those creatures are — at best — morally neutral. I would argue, and will continue to argue, that morality evolved and reflects the history of life and the physical environments of earth as the oft evolved eye reflects the radiative characterstics of the sun and also the optical characteristics of the earth’s atmosphere. I think that Gigerenzer must hold a similar position.
Returning to Gigerenzer’s ideas, he also says this in the introduction:
Social environments foster different strategies than physical environments, such as imitation instead of deliberation, and demand attention to information that is unique to social interaction, such as cues that could reveal one is being cheated or bluffed. [page 199]
I would qualify this a bit. I think that we might do well by adopting the attitudes of old-fashioned farmers, including the Amish, and coming into a communion of sorts with the parts of this world over which we are stewards. We interact with our physical environments and even blend into them in a sense. Though there are some short-term advantages to acting as if we were apart from our physical environments and trying to control them in the way we control pseudo-objects in a video game, we in the early 21st century are maybe seeing hints that our physical environments will take revenge of a sort when we act and think as if our surroundings were just a collection of objects independent of us and subject to our control. Given our tendencies towards unified thought and action, however weak and defective, an overlord attitude towards our physical environments may also spill over into our social and political attitudes, but the more important point is that we interact with our environments rather than standing above or outside them.
Our ways of thought towards our physical environments should reflect our true situation and I think some peoples, such as the Amish, have shown they can respect the trees on their land without turning them into ents or other magically conscious beings. As I understand it, evolutionary biology has supported the anthropological claim, based upon observation of foraging or pastoral societies, that human males in particular are totemic — they are very good at learning how to ‘think’ as if they were deer or bear or fish. Totemism will lead to magic rather than reasonable moral behavior, but we Christians at least have better possibilities open to us.
[S]ome have proposed mental modules for intuitive physics, mathematics, and biology — a view that turns academic subjects into domains. An evolutionary perspective suggests that a different division of labor has evolved, one directed at solving important adaptive problems, such as attachment development, mate search, parenting, social exchange, coalition formation, and maintaining and upsetting dominance hierarchies. …
The idea of modules, specialized for certain adaptive problems conflicts with the compartmentalization of psychology. Today’s areas of specialization are defined in terms of faculties, such as memory, thinking, decision making, intelligence, motivation, and emotion. These faculties have become institutionalized in modern university curricula and grant agencies. [page 200]
Unfortunately, they’ve also become institutionalized in our basic anthropologies as embedded in our theology and philosophy and history and literature. And, yet, I wouldn’t fully jettison the idea that man can abstract across a range of domains to discover some rules of thought. I don’t believe Gigerenzer intends to do so, but some enthusiastic followers might forget that an analysis such as this will correct exaggerations of the importance of abstract thought by using abstract thought as a tool. A balanced view is contained in the claim:
Adaptive problems and their modern equivalents, such as foraging and dieting and social exchange and markets, demand the orchestration of these faculties [such as memory and thinking], and not their segregation. [page 200]
A couple of pages later, he says: “I am not arguing against axioms and rules, only against their a priori imposition as context-independent yardsticks of rationality.”
Before reading Gigerenzer, I’d made similar claims. If we are to have any chance of solving our problems, we have to return to our roots, to understand human nature especially in its social aspects. From there, we can enlarge our viewpoint to maybe find ways to live moral lives in our huge, cosmopolitan societies.
In my worldview, axioms and rules — even those of mathematics — are tools set inside of factual stories. They are relationships between specific islands of order which God has raised up from an absolutely infinite ocean of facts. Ultimately, those islands are a setting for a story God is telling, the story which is this world — the universe seen in light of God’s purposes. This means that even the most elegant and abstract of mathematics has a sort of ad-hoc nature which can be seen in the multitude of geometries and algebras which seem necessary to describe the current state of the description of physical reality by mathematical physicists (see “The Road to Reality” by Roger Penrose). The real point I’d make is that the islands are important in being our homes or vacation spots but they’re insignificant in many ways relative to the surrounding oceans of factual reality, that is — chaos. The truly difficult tasks of human thought and human adaption to reality deal with those oceans which are factual and, hence, seem chaotic to human eyes and human minds.
In Chapter 9, Rationality: Why Social Context Matters, Gigerenzer takes on the goals of rational thought as advocated by modern economists and other social thinkers (he gives Samuelson the Nobel Laureate economist as on example). He provides examples which cast doubt upon the possibility of applying consistency or maximization rules when, for example, social context makes relevant additional choices which are logically irrelevant.
One interesting aspect of decision-making, made by at least 1939 by the psychologist Brunswik, is that animals make certain types of decisions as if they were maximizing in the context of a group. Many scientists assume that because they think of laboratory rats as individuals, those over-bred rodents should act as individuals or else they’re being irrational. In particular, if a rat is put in a maze where food is found with 80% probability with a left turn or 20% with a right turn, the rat will turn left only 68% of the time after learning. That rat is acting as if there are other rats looking for food. In that case, a willingness to take greater risks, to take the more difficult road, might well pay off.
The biological example given by Gigerenzer involves coloration where, say, white is usually a better camouflage color because of snow but some years have snow-less winters and black is better. The breeding population minimizes chances of being wiped out if there is always a certain percentage of individuals which have the black color though they will usually be eaten — thus modern theories of rationalization, if used by nature, might lead more easily to the extermination of such a breeding population.
Chapter 10, Domain Specific Reasoning: Social Contracts and Cheating Detection, starts with a recounting of John Garcia’s discovery that rats are not general purpose learning machines. They are wired, so to speak, in some fairly specific ways. His first finding was that rats didn’t readily learn to avoid flavored water if it was followed by an electric shock but they did if nausea was induced, even if the nausea didn’t hit until 2 hours later. It’s easier for rats, and humans as well, to associate tastes with stomach upsets than with electrical shocks.
I’ll digress just a little to puncture some of the conceits of modern scientists or their fans. John Garcia was the man who proved that B.F. Skinner was wrong and that our ‘conditioning’ is specific and not that of ‘stimulis equipotentiality’. For the crime of proving wrong a theory popular with the powers-that-be in the field of cognitive research, Garcia was censored from 1965 to 1979 or so. His papers were rejected by the editors of the journals of the American Psychological Association. Many of those editors were probably the sorts of yahoos who liked to joke about the closed-mindedness of those priests back in the Middle Ages. Every age has its forms of censorship which are exercised by those whose major asset is control of existing institutions.
Moving on to an example explored by the ‘evolutionary’ psychologist Leda Cosmides, Gigerenzer discusses experiments and evolutionary style reasoning that indicates that human beings confronted with a logical puzzle don’t generally reason well. I’ll not discuss the set-up of this particular experimental test, but I will say that I first let myself answer intuitively and came up with one of the ‘wrong’ possibilities and then I thought about the puzzle more explicitly and came up with the right answer — I was a math major in college and took logic courses and have done a fair amount of casual computer programming in the decades since. Tests with puzzles that imply a social contract with benefits and the possibility of cheating, reveal that human beings will reason as if to detect cheating on the part of test-givers or some third-party described in the set-up of the puzzles. We human beings are altruistic but will try to determine whether others around us are reciprocating or whether they take benefits without reciprocating. Again, this is hardly surprising on the part of social animals which do best in cooperative groups. Our cooperative relationships are far more important to us than formal reasoning — in general. At the same time, we have to remember most human beings are capable of some degree of formal reasoning and some, such as Einstein, are capable of extraordinary achievements in formal reasoning. We have to also remember that abstract reasoning is far more important in a complex world in which we often deal with situations or with groups of human being which are not familiar to us.
In Chapter 11, The Modularity of Social Intelligence, Gigerenzer tells us, “Primates appear to manipulate social objects with more ease and sophistication than physical tools.”
In the rest of the chapter, he speculates on the ways in which a modular social intelligence could work. In particular he addresses the problem which confronts much evolutionary reasoning about complex organs or behavioral traits: how could such have arisen from selection processes? The specific answers aren’t clear but it’s well-known that physical or behavioral traits which originally served one purpose will often be applied to other purposes. A biochemical analogue is the body’s frugal use of certain complex chemicals to help neutralize poisons often found in foods we eat and also to help regulate moods by altering brain activity. Certain ‘old-fashioned’ anti-depressant drugs destroy the ability of the patient’s system to neutralize poisons found in aged cheeses and other foods and, thus, those who use those drugs have to avoid certain foods which are harmless to most human beings. By some sort of accidental process, a complex chemical already present in the mammal’s body, probably first in the digestive system, was used for another purpose.
While I think Gigerenzer is right that our thinking evolved in ‘modules’ directed to solving specific problems (distinguishing between friends and enemies, finding proper mates, etc.), I think there is, in fact, a faculty in human beings that corresponds to a domain-general intelligence: our ability to construct morally purposeful narratives. It’s those narratives which are the setting for domain-specific forms of intelligence in human beings. The most important question about human intelligence is not how individual modules operate but rather: is there a narrative, a morally purposeful story, both true and accessible to human minds? When we think, are we — at least potentially — dealing with truths, perhaps contingent truths of this universe? Alternatively, was Pilate right in questioning if there is any truth? In that second case, our intelligence would be no more than a pragmatic and ad-hoc way of surviving and reproducing.
If we are potentially dealing with truths, it would remain true that no particular human mind could fully and truly comprehend the truths of this universe, empirical truths ordered to purposes which can’t be seen clearly unless we could see the universe in its entirety through time and space, but we know that there might well be all the sorts of moral truths in our universe which the Enlightenment thinkers and most of their Christian opponents banished to metaphysical regions.