60 pages • 2 hours read
Steven PinkerHow The Mind Works
Nonfiction | Book | Adult | Published in 1997A modern alternative to SparkNotes and CliffsNotes, SuperSummary offers high-quality Study Guides with detailed chapter summaries and analysis of major themes, characters, and more.
Symbols & Motifs
Computation
Pinker sets up his argument of how the mind works by introducing the computational theory of mind. This theory argues that the mind works using computations that result in neurons firing, triggering other neurons that fire if they reach their threshold, and ultimately building to identifying objects in our environment or speaking a sentence. Each step is a computation, and the combination of all the computations creates the final, larger outcome. Computation is, therefore, the language of the book. As Pinker moves into larger concepts like relationships and emotions, he still uses a computation framework, even if he is not explicit about it. Relationships are discussed in terms of the percentage of genes shared between individuals. Decisions are “weighed” by individuals before choosing a route. Experience “adds up,” people adjust connection weights as the evidence “accumulates” for a given trait belonging to a category. This kind of language is how many people talk about these phenomena. The idea of computation pervades our discussion of the mind, how it works, and how we live as humans.
Homunculi
Pinker brings up the idea of homunculi originally as a criticism leveraged against the computational theory of mind. Critics have argued that little homunculi would have to exist in the brain, making decisions about whether something was furry or barked or was brown. A homunculus is defined as a very small human or humanoid creature. The criticism implies that if we needed tiny little humans to live in our brain to make our mind work, we would then need to explain how the homunculi work, leaving us with the same questions we had before.
Pinker argues that neurons are these homunculi. They are not tiny humanoid creatures, but they can do what the critics claim would require homunculi. Sensory neurons offer the most straightforward example. They can respond to external stimuli, being turned on or off simply by the light waves, sound waves, pressure, or chemicals that reach the nose or mouth from food. These real external stimuli will trigger the neurons that respond to them to fire. The connected neurons will receive stimulation indicating something is around. Enough of these neurons firing will stimulate the next level, which receives input from, for example, the light waves showing red and yellow, the smell indicating cheese and tomato sauce. The light waves will also show the item has boundaries indicating it’s a triangle. These inputs together indicate pizza, and the brain will identify the item as likely pizza.
The process described above starts with simple neurons that respond to a single input, and these neurons are connected to neurons that respond to multiple inputs, and those neurons can connect to other neurons that respond to even more inputs and have even more outputs. The result is an almost humanoid collection of neurons. They have at least some properties of the human mind, and they can produce some of what we associate with being human, like thoughts, words, and actions. When they are combined to form an entire brain with all possible inputs and all possible outputs, we have the fully human mind. These homunculi are invoked further as Pinker builds his argument, reminding us of the combinatorial nature of the human brain and that the homunculi are an apt metaphor for the information-processing units that make up the brain.
Goals
The importance of goals to the human mind cannot be understated. Having goals is repeatedly mentioned as a key piece of intelligence, of human existence, of emotions, of relationships, and many other topics. Without goals, the development of the massively complex, capable, and expensive brain isn’t worth it. If we don’t want anything out of life, we hardly need intelligence or significant information-processing capacity.
However, the fact that we have goals and take them for granted as key to life does not mean that a process like natural selection has goals or that another creature with goals would necessarily have goals like ours. Natural selection is not a living being, and if goals are required to be a living being, then natural selection is unlikely to have them. If natural selection had goals, it would also disprove Darwin’s theory about how it worked. Goals would imply knowledge of a desired end state, and yet the best end state will depend on the environment in which the organism lives. Unless natural selection is controlled by an all-knowing being, it can’t know the environments available across billions of years of existence.
Many criticisms of natural selection have made the mistake of assuming that the current human state was the goal or end state. This misconception leads to inaccurate assertions, such as that the complex abilities of the human mind had to come fully formed and couldn’t have evolved with intermediate steps because those intermediate steps would not have been useful. However, that assertion assumes that the environments of our ancestors were the same as our environment today, which is patently false. The assertion also assumes that every ability we have now would have been used in the same way in all its intermediate steps. Instead, it is likely the intermediate versions of the mind and other organs could easily have been used for other functions, as Pinker shows with the development of wings. Even when reading the book and attempting to keep in mind that natural selection doesn’t have goals, it is tempting to interpret human evolution and skill development as being goal-directed. It takes effort to continually develop explanations for evolution that avoid goals.
Humans have an easier time interpreting something with goals because of how central they are to our lives. We must ignore our basic operating mode and step outside it to consider something goal-less, something radically different from a human. We struggle even with organisms whose goals would likely be different from our own, but we can get closer and find it easier to simply imagine different goals. This issue with understanding something goal-less, therefore, speaks to one of the themes in this book that humans start with what they know and build new knowledge from there. We assume that something unknown that shares traits with something known will be similar in ways we haven’t observed. Having goals appears to be a base requirement in our reasoning. We apply this reasoning to everything unknown, and only with considerable evidence and effort will we allow something to be goal-less.
Related Titles
By Steven Pinker
