From the Introduction:
So there are two strands that intertwine in the notion of cosmopolitanism. One is the idea that we have obligations to others, obligations that stretch beyond those to whom we are related by the ties of kith and kind, or even the more formal ties of shared citizenship. The other is that we take seriously the value not just of human life but of particular human lives, which means taking an interest in the practices and beliefs that lend them significance. People are different, the cosmopolitan knows, and there is much to learn from our differences. Because there are so many human possibilities worth exploring, we neither expect nor desire that every person or every society should converge on a single mode of life. Whatever our obligations are to others (or theirs to us) they often have the right to go their own way. As we'll see, there will be times when these two ideals--universal concern and respect for legitimate difference--clash. There's a sense in which cosmopolitanism is the name not of the solution but of the challenge.
In geological terms, it has been a blink of an eye since human beings first left Africa, and there are few spots where we have not found habitation. The urge to migrate is no less "natural" than the urge to settle. At the same time, most of us who have learned the languages and customs of other places haven't done so out of mere curiosity. A few were looking for food for thought; most were looking for food. Thoroughgoing ignorance about the ways of others is largely a privilege of the powerful. The well-traveled polyglot is as likely to be among the worst off as among the best off--as likely to be found in a shantytown as at the Sorbonne. So cosmopolitanism shouldn't be seen as some exalted attainment: it bgeins with the simple idea that in the human community, as in national communities, we need to develop habits of coexistence...
From the chapter "The Escape from Positivism":
The deepest problem with Positivism, however, is not in its conclusions. It is in its starting point. I began, as I think one must if one is to make the Positivist story believable, with a single person, acting on her own beliefs and desires. Starting from there, one has to give an account of values that begins with what it is for me--this single person--to regard something as valuable. But to understand how values work, you must see them not as guiding us as individuals on our own but as guiding people who are trying to share their lives.
The philosopher Hilary Putnam famously argued that, as he once put it, "Meanings ain't in the head."...
We go astray, similarly, when we think of a moral vocabulary as the possession of a solitary individual. If meanings ain't in the head, neither are morals. The concept of kindness, or cruelty, enshrines a kind of social consensus. An individual who decides that kindness is bad and cruelty good is acting like Lewis Carroll's Humpty-Dumpty, for whom a word "means just what I choose it to mean--neither more, nor less." The language of values is, after all, language. And the key insight of modern philosophical reflection on language is that language is, first and foremost, a public thing, something we share. Like all vocabulary, evaluative language is primarily a tool we use to talk to one another, not an instrument for talking to ourselves.
From the chapter "Facts on the Ground":
In the early twentieth century, the French physicist Pierre Duhem noticed an interesting fact about the way scientists behave. When they do experiments or collect data to support their theories, other scientists, often those attached to different theories, deny that the evidence shows any such thing. The objections can be of many different kinds. They might say, for example, that the experiment really hasn't been done properly. (Your test tubes were contaminated.) They might say that the so-called data are simply incorrect. (We did the same experiment, and that's not what happened.) Or they could point out that their own theory explained the data just as well (The theory that life on Earth arrived in the form of basic organisms on a meteorite explains the fossil data just as well as the theory that life evolved by the creation of its basic elements as a result of electrochemical processes in the primeval oceans.) Starting with this observation, he went on to propose a general claim that philosophers know as the Duhem thesis. However much data you have, Duhem said, there will be many theories that explain it equally well. Theories, to use the jargon, are underdetermined by the evidence.
Underdetermination is worrying enough. But a later student of scientific thinking, the philosopher N. R. Hanson, noticed something equally troubling for the Positivist view about scientific thinking....What Hanson noticed was that the data never came free of theoretical commitments. When Galileo said that he saw through the telescope that the moon had mountains, he was assuming--as some of his opponents at the time pointed out--that telescopes work just as well in space as on Earth. That happens to be right. But how did he know? No one, at that point, had ever taken a telescope up into space to check. He just theorized that it was so. And, in fact, it turns out to be enormously difficult--Hanson thought it was literally impossible--to present data in language that isn't infused with theoretical ideas.
From the chapter "The Primacy of Practice":
...our political coexistence, as subjects or citizens, depends on being able to agree about practices while disagreeing about their justification. For many long years, in medieval Spain under the Moors and later in the Ottoman Near East, Jews and Christians of various denominations lived under Muslim rule. This modus vivendi was possible only because the various communities did not have to agree on a set of universal values. In seventeenth-century Holland,...the Sephardic Jewish community began to be increasingly well integrated into Dutch society, and there was a great deal of intellectual as well as social exchange between Christian and Jewish communities. Christian toleration of Jews did not depend on express agreement on fundamental values.
...let's recognize this simple fact: a large part of what we do we do because it just what we do. You get yp in the morning at eight-thirty. Why that time? You have coffee and cereal. Why not porridge? You send kids to school. Why not teach them at home? You have to work. Why that job, though? Reasoning--by which I mean the public act of exchanging stated justifications--comes in not when we are going on in the usual way, but when we are thinking about change. And when it comes to change, what moves peple is often not an argument from a principle, not a long discussion about values, but just a gradually acquired new way of seeing things.
Or consider another example: in much of Europe and North America, in places where a generation ago homosexuals were social outcasts and homosexual acts were illegal, lesbian and gay couples are increasingly being recognized by their families, by society, and by the law. This is true despite the continued opposition of major religious groups and a significant and persistent undercurrent of social disapproval. Both sides make arguments, some good, most bad, if you apply a philosophical standard of reasoning. But if you ask the social scientists what has produced this change, they will rightly not start with a story about reasons. They will give you a historical account that concludes with a sort of perspectival shift. The increasing presence of "openly gay" people in social life and in the media has changed our habits. Over the last thirty or so years, instead of thinking about the private activity of gay sex, many Americans started thinking about the public category of gay people. Even those who continue to think of the sex with disgust now find it harder to deny these peple their respect and concern...
Now I don't deny that all the time, at every stage, peple were talking, giving each other reasons to do things: accept their children, stop treating homosexuality as a medical disorder, disagree with their churches, come out. Still, the short version of the story is basically this: people got used to lesbians and gay people. I am urging that we should learn about people in other places, take an interest in their civilizations, their arguments, their errors, their achievements, not because that will bring us to agreement, but because it will help us get used to one another.
I am a philosopher. I believe in reason. But I have learned in a life of university teaching and research that even the cleverest people are not easily shifted by reason alone--and that can be true even in the most cerebral of realms. One of the great savants of the postwar era, John von Neumann, liked to say, mischievously, that "in mathematics you don't understand things, you just get used to them." In the larger world, outside the academy, people don't always even care whether they seem reasonable. Conversation, as I've said, is hardly guaranteed to lead to agreement about what to think and feel. Yet we go wrong is we think the point of conversation is to persuade, and imagine it proceeding a a debate, in which points are scored for the Proposition and the Opposition. Often enough, as Faust said, in the beginning is the deed: practices and not principles are what enable us to live together in peace. Conversations across boundaries of identity--whether national, religious, or something else--begin with the sort of imaginative engagement you get when you read a novel or watch a movie or attend to a work of art that speaks from someplace other than your own. So I'm using the word "conversation" not only for literal talk but also as a metaphor for engagement with the experience and the ideas of others. And I stress the role of the imagination here because the encounters, properly conducted, are valuable in themselves. Conversation doesn't have to lead to consensus about anything, especially not values; it's enough that it helps people get used to one another.