Theory and Reality

by Peter Godfrey-Smith

  What is Popper's single most important and enduring contribution to philosophy of science? I'd say it is his use of the idea of "riskiness" to describe the kind of contact that scientific theories have with observation. Popper was right to concentrate on the ideas of exposure and risk in his description of science. Science tries to formulate and handle ideas in such a way that they are exposed to falsification and modification via observation. Popper's formulation is valuable because it captures the idea that theories can appear to have lots of contact with observation when in fact they only have a kind of "pseudo-contact" with observation because they are exposed to no risks. This is an advance in the development of empiricist views of science. Popper's analysis of how this exposure works does not work too well, but the basic idea is good.

  4.6 Further Comments on the Demarcation Problem

  Popper is onto something when he says that scientific theories should take risks. In this section I will try to develop this idea a bit differently.

  Popper was interested in distinguishing scientific theories from unscientific ones, and he wanted to use the idea of risk-taking to make the distinction. But this idea of risk-taking is better used as a way of distinguishing scientific from unscientific ways of handling ideas. And we should not expect a sharp distinction between the two.

  The scientific way of handling an idea is to try to connect it with other ideas, to embed it in a larger conceptual structure, in a way that exposes it to observation. This "exposure" is not a matter of simple falsification; there are many ways in which exposure to observation can be used to modify and assess an idea. But if a hypothesis is handled in a way that keeps it apart from all the risks associated with observation, that is an unscientific handling of the idea.

  So it is a mistake to try to work out whether theories like Marxism or Freudianism are themselves "scientific" or not, as Popper did. A big idea like Marxism or Freudianism will have scientific and unscientific versions, because the main principles of the theory can be handled scientifically or unscientifically. Scientific versions of Marxism and Freudianism are produced when the main principles are connected with other ideas in a way that exposes these principles to testing. To scientifically handle the basic principles of Marxism is to try to work out what difference it would make to things we can observe if the Marxist principles were true. To do this it is not necessary that we write down some single observation that, if we observe it, will lead us to definitively reject the main principles of the theory. It will remain possible that an auxiliary assumption is at fault, and there is no simple recipe for adjudicating such decisions.

  To continue with Popper's examples, Marxism holds that the driving force of human history is struggle between economic classes, guided by ongoing changes in economic organization. This struggle results in a predictable sequence of political changes, leading eventually to socialism. Freudianism holds that the normal development of a child includes a series of interactions and conflicts between unconscious aspects of the child's mind, where these interactions have a lot to do with resolving sexual feelings toward his or her parents. Adventurous ideas like these can be handled scientifically or unscientifically. Over the twentieth century, the Marxist view of history has been handled scientifically enough for it to have been disconfirmed. Too much has happened that seems to have little to do with class struggle; the ever-increasing political role of religious and cultural solidarity is an example (Huntington 1996). And capitalist societies have adapted to problems-especially economic tensions-in ways that Marxist views about politics and economics do not predict. Of course, it remains possible to hang onto the main principles of Marxism despite this, but fewer and fewer people handle the theory in that way anymore. Many still think that Marxism contains useful insights about economic matters, but the fundamental claims of the theory have not stood up well.

  Freudianism is another matter; the ideas are still popular in some circles, but not because of success under empirical testing. Instead, the theory seems to hang around because of its striking and intriguing character, and because of a subculture in fields such as psychotherapy and literary theory which guards the main ideas and preserves them despite their empirical problems. The theory is handled very unscientifically by those groups. Freud's theory is not taken seriously by most scientifically oriented psychology departments in research universities, but it is taking a while for this fact to filter out to other disciplines.

  Evolution is another big idea that can be handled scientifically or unscientifically. People (including Popper) have wondered from time to time whether evolutionary theory, or some specific version of it such as Darwinism, is testable. So they have asked, What observations would lead scientists to give up current versions of evolutionary theory? A one-line reply that biologists sometimes give to this question is "a Precambrian rabbit." An evolutionary biology textbook by Douglas Futuyma expresses the same point more soberly: finding "incontrovertibly mammalian fossils in incontrovertibly pre-cambrian rocks" would "refute or cast serious doubt on evolution" (1998, 760). The one-liner is a start, but the real situation is more complicated. So let us look at the case.

  The Precambrian era ended around 540 million years ago. Suppose we found a well-preserved rabbit fossil in rocks boo million years old. All our other evidence suggests that the only animals around then were sponges and a few other invertebrates and that mammals did not appear until over 300 million years later. Of course, a good deal of suspicion would be directed toward the finding itself. How sure are we that the rocks are that old? Might the rabbit fossil have been planted as a hoax? Remember the apparent fossil link between humans and apes that turned out to be a hoax, the Piltdown man of 1908 (see Feder 1996). Here we encounter another aspect of the problem of holism about testing-the challenging of observation reports, especially observation reports that are expressed in a way that presupposes other pieces of theoretical knowledge. This will be discussed in chapter io. But let us suppose that all agree the fossil is clearly a Precambrian rabbit.

  This finding would not be an instant falsification of all of evolutionary theory, because evolutionary theory is now a diverse package of ideas, including abstract theoretical models as well as claims about the actual history of life on earth. The theoretical models are intended to describe what various evolutionary mechanisms can do in principle. Claims of that kind are usually tested via mathematical analysis and computer simulation. Smallscale evolution can also be observed directly in the lab, especially in bacteria and fruit flies, and the Precambrian rabbit would not affect those results.

  But a Precambrian rabbit fossil would show that somewhere in the package of central claims found in evolutionary biology textbooks, there are some very serious errors. These would at least include errors about the overall history of life, about the kinds of processes through which a rabbitlike organism could evolve, and about the "family tree" of species on earth. The challenge would be to work out where the errors lie, and that would require separating out and independently reassessing each of the ideas that make up the package. This reassessment could, in principle, result in the discarding of very basic evolutionary beliefs-like the idea that humans evolved from nonhumans.

  Over the past twenty years or so, evolutionary theory has in fact been exposed to a huge and sustained empirical test, because of advances in molecular biology. Since the time of Darwin, biologists have been trying to work out the total family tree linking all species on earth, by comparing their similarities and differences and taking into account factors such as geographical distribution. The family tree that was arrived at prior to the rise of molecular biology can be seen summarized in various picturesque old charts and posters.

  Then more recently, molecular biology made it possible to compare the DNA sequences of many species. Similarity in DNA is a good indicator of the closeness of evolutionary relationship. Claims about the evolutionary relationships between different species can be tested reasonably directly by discovering how similar their DNA is and calcul
ating how many years of independent evolution the species have had since they last shared a "common ancestor." As this work began, it was reasonable to wonder whether the wealth of new information about DNA would be compatible or incompatible with the family tree that had been worked out previously. Suppose the DNA differences between humans and chimps had suggested that the human lineage split off from the lineage that led to chimps many hundreds of millions of years ago and that humans are very closely genetically related to squid. This would have been a disaster for evolutionary theory, one of almost the same magnitude as the Precambrian rabbit.

  As it happened, the DNA data suggest that humans and chimps diverged about 4.6-5 million years ago and that chimps or pigmy chimps (bonobos) are our nearest living relatives. Prior to the DNA data, it was unclear whether humans were more closely related to chimps or to gorillas, and the date for the chimp-human divergence was much less clear. That is how the grand test of our old pre-molecular family tree has tended to go. There have been no huge surprises but lots of new facts and a lot of adjustments to the previous picture.

  Further Reading

  Popper's most famous work is his book The Logic of Scientific Discovery, published in German in 193 5 and in English in 1959. The book is mostly very readable. Chapters 1-5 and 1o are the key ones. For the issues in section 4.4 above, see chapter 5 of Popper; for section 4.5, see chapter 1o. A quicker and very useful introduction to Popper's ideas is the paper "Science: Conjectures and Refutations" in his collection Conjectures and Refutations (1963).

  Newton-Smith, The Rationality of Science (1981), contains a clear and detailed assessment of Popper's ideas. It includes a simplified presentation of some of the technical issues surrounding corroboration that I omitted here. Salmon 1981 is an exceptionally good critical discussion of Popper's views on induction and prediction. See also Putnam 1974. Schilpp (1974) collects many critical essays on Popper, with Sir Karl's replies.

  Popper's influence on biologists and his (often peculiar) ideas about evolutionary theory are discussed in Hull 1999. Horgan's book The End of Science (1996) contains a very entertaining interview with Popper.

  5.1 "The Paradigm Has Shifted"

  In this chapter we encounter the most famous book about science written during the twentieth century-The Structure of Scientific Revolutions, by Thomas Kuhn. Kuhn's book was first published in z96z, and its impact was enormous. Just about everything written about science by philosophers, historians, and sociologists since then has been influenced by it. The book has also been hotly debated by scientists themselves. But Structure (as the book is known) has not only influenced these academic disciplines; many of Kuhns ideas and terms have made their way into areas like politics and business as well.

  A common way of describing the importance of Kuhn's book is to say that he shattered traditional myths about science, especially empiricist myths. Kuhn showed, on this view, that actual scientific behavior has little to do with traditional philosophical theories of rationality and knowledge.

  There is some truth in this interpretation, but it is often greatly exaggerated. Kuhn spent much of his time after Structure trying to distance himself from some of the radical views of science that came after him, even though he was revered by the radicals. And the connection between Kuhns views and logical empiricism is actually quite complicated. For example, it comes as a surprise to many to learn that Kuhns book was published in a series organized and edited by the logical empiricists; Structure was published as part of their International Encyclopedia of Unified Science series. As a matter of historical fact, though, there is no denying that this was something of a "Trojan horse" situation. Logical empiricism was widely perceived as being seriously damaged by Kuhn.

  I said above that some of Kuhns ideas and terms have made their way into areas far from the philosophy of science. The best example is Kuhn's use of the term "paradigm." Here is a passage from Tom Wolfe's 1998 novel, A Man in Full. Charlie Croker, a real estate developer who has debt problems, is talking with his financial adviser, Wismer ("Wiz") Stroock.

  "I'm afraid that's a sunk cost, Charlie;' said Wismer Stroock. "At this point the whole paradigm has shifted."

  Charlie started to remonstrate. Most of the Wiz's lingo he could put up with, even a "sunk cost." But this word "paradigm" absolutely drove him up the wall, so much so that he had complained to the Wiz about it. The damned word meant nothing at all, near as he could make out, and yet it was always "shifting," whatever it was. In fact, that was the only thing the "paradigm" ever seemed to do. It only shifted. But he didn't have enough energy for another discussion with Wismer Stroock about technogeekspeak. So all he said was:

  "OK, the paradigm has shifted. Which means what?" (71)

  This sort of talk derives completely from Kuhn. But what is a paradigm? The short answer is that a paradigm, in Kuhn's theory, is a whole way of doing science, in some particular field. It is a package of claims about the world, methods for gathering and analyzing data, and habits of scientific thought and action. In Kuhns theory of science, the big changes in how scientists see the world-the "revolutions" that science undergoes every now and then-occur when one paradigm replaces another. Kuhn argued that observational data and logic alone cannot force scientists to move from one paradigm to another, because different paradigms often include within them different rules for treating data and assessing theories. Some people have interpreted Kuhn as claiming that changes between paradigms are completely irrational, but Kuhn definitely did not believe that. Instead, Kuhn had a complicated and subtle view about the roles of observation and logic in scientific change.

  In a passage like the Tom Wolfe one above, "paradigm" is used in a looser way derived from its role in Kuhn's theory of science. A paradigm in this sense means something like a way of seeing the world and interacting with it.

  Kuhn did not invent the word "paradigm." It was an established term, which meant (roughly) an illustrative example of something, on which other cases can be modeled. Kuhn discusses this original meaning in Structure (1996, 2..3). And although Kuhns theory is the inspiration for all the talk about paradigm shifts that one hears, Kuhn only occasionally used the phrase "paradigm shift." More often he talked about paradigms changing or being replaced. Whichever term one uses, though, Kuhns theory was itself something like a paradigm change in the history and philosophy of science. Nothing has been the same since.

  5.2 Paradigms: A Closer Look

  A moment ago I said that a paradigm in Kuhn's theory is a package of claims about the world, methods for gathering and analyzing data, and habits of scientific thought and action. However, it is more accurate to say that this is one sense in which Kuhn used the term "paradigm." In Structure, the term is used in several different ways; one critic counted as many as twenty-one different senses (Masterman 1970). Kuhn later agreed that he had used the word ambiguously, and throughout his career he kept finetuning this and other key concepts. To keep things simple, though, in this book I will recognize two different senses of the term "paradigm."

  The first sense, which I will call the broad sense, is the one I described above. Here, a paradigm is a package of ideas and methods, which, when combined, make up both a view of the world and a way of doing science. When I say "paradigm" in this book without adding "broad" or "narrow," I mean this broad sense. But there is also a narrower sense. According to Kuhn, one key part of a paradigm in the broad sense is a specific achievement, or an exemplar. This achievement might be a strikingly successful experiment, such as Mendel's experiments with peas, which eventually became the basis of modern genetics. It might be the formulation of a set of equations or laws, such as Newton's laws of motion or Maxwell's equations describing electromagnetism. Whatever it is, this achievement is a source of inspiration to others; it suggests a way to investigate the world. Kuhn often used the term "paradigm" just for a specific achievement of this kind. I will call these achievements paradigms in the narrow sense. So paradigms in the broad sense (whole ways of doing s
cience) include within them paradigms in the narrow sense (examples that serve as models, inspiring and directing further work). Kuhn himself did not use this "narrow/broad" terminology, but it is helpful. When Kuhn first introduced the term "paradigm" in Structure, he defined it in the narrower sense. But in much of his writing, and in most of the work written after Structure using the term, the broad sense is intended.

  Kuhn used the phrase "normal science" for scientific work that occurs within the framework provided by a paradigm. A key feature of normal science is that it is well organized. Scientists doing normal science tend to agree on which problems are important, on how to approach these problems, and on how to assess possible solutions. They also agree on what the world is like, at least in broad outlines. A scientific revolution occurs when one paradigm breaks down and is replaced by another.

  This initial sketch is enough for us to go straight to some central points about the message of Kuhns book.

  The first point can be approached via a contrast with Popper. For Popper, science is characterized by permanent openness, a permanent and allencompassing critical stance, even with respect to the fundamental ideas in a field. Other empiricist views will differ on the details here, but the idea of science as featuring permanent openness to criticism and testing is common to many versions of empiricism. Kuhn disagreed. He argued that it is false that science exhibits a permanent openness to the testing of fundamental ideas. Not only that, but science would be worse off if it had the kind of openness that philosophers have treasured.

  The second point concerns scientific change. Here again a contrast with Popper is convenient. For Popper, all science proceeds via a single process, the process of conjecture and refutation. There can still be episodes called "revolutions" in such a view, but revolutions are just different in degree from what goes on the rest of the time; they involve bigger conjectures and more dramatic refutations. For Kuhn, there are two distinct kinds of scientific change: change within normal science, and revolutionary science. (These are bridged by "crisis science;' a period of unstable stasis.) These two kinds of change have very different epistemological features; when we try to apply concepts such as justification, rationality, and progress to science, according to Kuhn we find that normal and revolutionary science have to be described very differently. Within normal science, there are clear and agreed-upon standards for the justification of arguments; within revolutionary science there are not. Within normal science there is clear progress; within revolutionary science it is very hard to tell (and it is hard to even interpret the question). Because revolutions are essential to science, the task of describing rationality and progress in science as a whole becomes very complicated.