Summary
Rosenberg sets out to ‘prove’ through Neuroscience that the way we understand our past, present, and future might not be based on a misunderstanding. In what is sure to ruffle the feathers of academics of every stripe, Rosenberg uses various studies as a lever to overturn several common theories of mind. For the uninitiated, a theory of mind is an explanatory framework whose purpose is to explain the mind to itself. Most common theories of mind rely on the iconic duo of desire and action. Charles is crying because he can’t get an ice cream cone. In the previous sentence, we are met with Charles’ desire and the action that results from the desire, a cause and an effect. Rosenberg then uses this as a jumping-off point to argue that this core assumption that almost all theories of mind make is flawed.
This is a tall order because how else can we understand Charles’ crying if not by its dependent relation with the ice cream cone, the object of his desire? To understand Rosenberg’s argument, the first step is to realize that an effect can have multiple seemingly plausible explanations. For example, you could go the crank psychologist route: Charles is crying because he thinks he wants the ice cream cone, but in reality, the ice cream cone is a substitute for his mother’s affection. Another possible explanation: a KGB spy injected Charles with a serum that causes him to cry when he sees ice cream. The important thing to remember is that causes and effects are linked by a story we construct, and the story may or may not be true.
I think one of the best ways to conceptualize the possibility that Rosenberg is pushing on his readers is to imagine a tortoise wandering around an island doing tortoise things, but all the while the tortoise is followed closely by a hummingbird. This hummingbird narrates all the tortoise’s actions so the observer can understand the causes. The longer the hummingbird follows the tortoise, the better it becomes at anticipating the behavior patterns of the tortoise, so much so that the narration mostly coincides with the actions taken by the tortoise. Rosenberg would argue this scenario is analogous to the condition we find ourselves in. Our bodies are finely tuned machines responding to various stimuli in the external environment. At the same time, we have this rich conscious experience that runs a constant commentary on what it thinks our body is up to. The key insight here is to realize that this implies that conscious experience is retroactive—that is, our experience is a post-hoc explanation of an event that has already happened, and just like with the hummingbird, our explanations may or may not coincide with the ‘reality’ of the actions that our body is taking. This is easy enough to intuit when it comes to autonomic processes, like when your stomach growls, you may experience it as your body being hungry but would also be willing to adopt a different explanation if one were offered by your doctor. This is a harder pill to swallow when it comes to conscious centric experiences like anger, rationality, or awe. At this point, the inner monologue is so compelling that it becomes almost impossible not to believe the “reasons” our mind generates for feeling one way or another. Rosenberg would argue that in reality the line between autonomic and voluntary processes don’t exist. The difference is just in how convincing our inner voice is.
If this is true, then the title starts to make much more sense. At the same time, it reveals how the title is, in many ways, too narrow. If we are inscrutable to ourselves how much more so must others be? This would make history, social sciences, psychology, and many more pursuits to understand human actions no more than a fool’s errand.
Thoughts
A provocative title for a provocative book, it is somewhat reminiscent of the ever present “You’re doing ‘X’ wrong” articles where ‘X’ is something like using a can opener. Overall, the arguments in the book are compelling. I am not qualified to comment on the neuroscientific research or conclusions that are in the book, but then again, one could argue that Rosenberg isn’t either. One must wonder, why if the conclusions from neuroscience are as clear as Rosenberg claims them to be, the opinions of neuroscientists haven’t more closely clustered around Rosenberg’s ideas. Rosenberg would argue this reflects how deeply rooted our intuitions about theory of mind are. On the other hand, I would argue that it’s also plausible his conclusions aren’t the only valid interpretations of the data.
While this is true, advances in language models seem to be aligning with some of Rosenberg’s intuitions. Most interestingly, it has been demonstrated that when you ask an LLM to explain how it reached its conclusions, you can’t just accept its explanation. The reason is that the explanation it generates may be a hallucination. The parallels between the architecture of LLM’s and the way human’s think points in the same direction as Rosenberg’s retroactive explanation. Rosenberg compares himself to fringe thinkers like Darwin and Copernicus, arguing that his ideas will eventually gain acceptance as evidence accumulates. I also agree that the direction of research tends to be moving towards his position, but I also think this book lacks the clarity to be the “Origin of Species” of a new movement. Whether or not Rosenberg’s opinions become mainstream is something only time will tell, and hopefully they don’t become another thing that history gets wrong.