I’ve finally finished Steven Pinker’s wonderful new book The Stuff of Thought. The question it poses is deceptively simple: what can the way we use language tell us about the way the mind works? The investigation, laid out in nearly 500 pages of sparkling prose, takes us from verb forms, to Jerry Fodor’s outrageous “extreme nativism”, to the vocabularies of space and time, to Bill Clinton’s testimony, to the metaphor metaphor, to Kripke’s “rigid designators”, to swearing, to bribing the maitre d’. But the result of this diversity is a remarkably simple, coherent, and plausible theory that describes how human beings manage the collection of thoughts which make up our models of the world and our relationships to others.
Pinker’s conclusions are summarized in five pages (428-432) of the last chapter of his book. I found myself re-reading the chapter several times, with highlighter in hand, and eventually I typed up the phrases that I’d highlighted and knitted them together into a “summary of the summary”. I’m including it below the fold. All the memorable words are Pinker’s; the crude packaging is mine.
One aspect of this model is its relationship to brain structure. Although he mentions many important experimental results in connection with individual steps in the investigation, Pinker downplays this when he comes to summarize his theory. Many of its elements are already verified by experimental work, and the functional level of the model is such that most of the more controversial or speculative bits should be susceptible to experimental [dis]confirmation over the next few years. This won’t stop the mysterians like Chalmers from pressing their forms of dualism, but it should give us a solid functional model for humans or zombies…
Although I talk about “Pinker’s theory”, it is obviously based on the work of many others. Nevertheless, I think I shall find myself referring, perhaps silently, to “the Pinker model” as I read about consciousness, psychology, philosophy of mind, and so forth.
—
Humans package their experience into objects and events, which they assemble into propositions. They are built out of a clear inventory of units: events, states, things, substances, places, and goals. Units can do things: acting, going, changing, being, having. An event may impinge on another, by causing, enabling or preventing it. An action may have a goal: a destination (of motion), or a state resulting from change. Objects are differentiated as human or nonhuman, animate or inanimate, solid or aggregate, and how they are laid out in space. Events take up time, and are ordered.
Humans recognize unique individuals as well as categories. Stable categories are essential and defining, as distinct from transitory or superficial ones. When humans think about an entity – where or what it is, or how it changes or moves – they treat it holistically, as a blob without internal parts. Objects move as a whole, are suffused by a trait, or change from one state to another entirely. But humans can also articulate an object into its parts and register how they are related. When the object is a human body, the concept of person emerges. A person is his body but also has his body parts, along with possessions including chattels, ideas, and fortune.
Humans have a primitive concept of number: one, two and many, although they can estimate larger quantities approximately. This coarse quantification applies to counting objects as well as locating things in space and events in time.
Humans do not reckon space in smooth coordinates; instead they impose a coordinate frame based on the earth, their bodies, or a prominent object, and use qualitative spatial relations. These distinguish up from down and front from back, and capture topological relations such as touching, inside, near, or attached to, but they are unreliable in distinguishing left from right. Spatial thinking is tailored to the demands of manipulating things, so it is not defined by geometry alone but to an intuitive physics of fitting supporting, containing, covering, and other ways that humans use things.
Humans divide time into a psychological present (~3 seconds long), an indefinite past, and an indefinite future. The past and future are often infected with metaphysics: the past is actual, the nonpast is hypothetical, and the future is willable.
Humans see some things as just happening and others as being caused. Causality is assessed not just by correlation or pondering alternatives, but by recognizing an impetus that is transferred from a potent, moving agent to a weaker, resisting entity. This in turn gives rise to intuitions of helping, hindering, preventing, and allowing. The first link in a causal chain is an action initiated by a (usually human) agent. Actions are defined by their manner, the change they bring about, or both. Humans care about whether change is brought about accidentally or intentionally, directly or via a proxy, and as a means or an end. They moralize these distinctions, holding agents blameworthy for the events they voluntarily, intentionally, and directly cause to happen.
Humans don’t just entertain ideas, but steep them with emotion. They are in awe of deities and the supernatural. They are terrified by disease, death, and infirmity. They are revolted by bodily secretions, and take a prurient interest in sexuality. They loathe enemies, traitors, and subordinate peoples. As unpleasant as these thoughts are, people will inflict them on one another, to intimidate, denigrate, attract attention, or demonstrate strength.
Humans are touchy about their relationships. They maintain a face which emboldens them to stake out claims in negotiation. They are sensitive to social rank, solidarity, and empathy. They distinguish between those with whom they share resources and feel ties of empathy, those with whom they negotiate status or power, and those with whom they trade goods and services.
Relationships are invested with a moral coloring, and people feel embarrassment when they breach the logic of a relationship; they feel contempt for those who breach it deliberately. Thus humans are vulnerable to public acknowledgement of such an act, but often risk these breaches, sometimes to renegotiate a relationship. As a result, they engage in hypocrisy and taboo, designed to preserve the mutual knowledge that maintains a relationship even as they transact business inconsistent with it.
Ugh. Re-reading this, I’m appalled at how I’ve butchered Pinker’s language. Nevertheless, I hope it’s useful.