A delightful polemic with a valuable point. Donald dramatically uses the intricate demands of a face to face conversation to show the practical weaknesses in the laboratory view of short and long term memory. The laboratory evidence that working memory is very limited is overwhelming, and has fed the modern philosophical trend toward viewing conscious awareness as an illusory result of the work of unconscious agents.
But things we do in daily life clearly require us to track things much more numerous and much longer than could possibly be accomplished by "seven plus or minus two" chunks, even with clever strategies for grouping things. Donald uses this to argue that conscious processes are very real and not to be ignored, and do play a central role in human intelligence.
Donald unflinchingly takes on the likes of "hardliners" such as Dan Dennett who argue that there is no central "meaner," no self, no little person in our heads observing the stream of consciousness in a Cartesian theater. He points out that the drafts we generate in our minds are not at all arbitrary competitors for dominance, but are distinctly related to goals and expectations. Most insightfully, he argues that discounting the role of conscious processes has dire implications for social and political philosophy and how we view human responsibility for our own actions.
In my view, Donald makes the excellent point for yet poorly understood intermediate term memory mechanisms very convincingly. I was completely persuaded that this is something we need to study to understand human abilities, and that "hardliners" views have some weaknesses I hadn't considered seriously before. He does make one rhetorical twist, though, that confused and sometimes annoyed me until I figured it out. He argues convincingly that we should retain the ideas of executive processes, goals, schema, and expectations, and how they influence thinking. The mind is organized in a central and domain general way for many critical things, rather than being completly modular and the result of bottom-up processing by independent functional agents.
I bought his argument here from fairly early in this excellent book. But then he also consistently equates this kind of organization with what other people call "consciousness," without making it clear at first. So you start wondering why he is calling all sorts of things "conscious" when clearly we
don't notice them !
Most strikingly, in reviewing the research on subliminal effects, he considers them conscious, even though they are seemingly by definition, not ! That is where I discovered that he is relating conscious processes to goal direction and selective attention, not to "noticing." "Noticing" per se actually has very little to do with anyghing in this book. This was a difficult conceptual turn for me, but may be a profound idea. It preserves the idea of consciousness as the selective goal-oriented use of attention to organize the activity of the mind, but doesn't attempt to explain phenomenal awareness per se. His idea of the substrate of consciousness is a neccessary but not sufficient basis for "noticing." The emphasis here is on how we select things to focus our resources on, rather than how phenomenal experience arises. This shift of emphasis allows him to make short work of some of the paradoxical ideas of the hardliners, without trying to tackle the "hard questions" of consciousness directly.
In a way, Merlin Donald takes on the role toward the study of the human mind that Gould, Lewontin, and Rose take toward the study of human evolution. He tries valiantly to bring us back from what he sees as the brink of an awful and unwarranted reductionism. The reductionism of mind to unconscious computation, he points out, threatens the very foundation of our political and economic ideas around freedom and individual responsibility.
Remarkably enough, I think his argument often succeeds.
One of the reasons his argument succeeds is that he makes a very clear distinction between limited consciousness and non-existent consciousness, a line that gets blurred by some philosophers in the process of trying to explain subjective experience in terms of neurons.
Donald describes the difference that makes a difference, that human beings can select their own goals and adjust their own priorities because their nervous system is patterned by a symbolic web of culture to form a distributed cognitive network. Going directly against the modern trend of evolutionary psychology in explaining away awareness as an artifact of functional computational modules, the author argues that human minds do have one very important distinction from other primate minds, a unique additional capacity for consciousness that evolved from the unique conditions of human evolution. The human mind is not, he suggests, simply the result f emergent qualities of an arbitrarily complex neural network. That would be too glib an explanation, and wouldn't explain why sensory nets are aware and not motor nets.
This book seems to be a manifesto of sorts toward a new view of the mind that incorporates what we know about the self and goal-directed domain-independent behavior rather than explaining away these important aspects of human mental function.