Crop circles are geometrical patterns of flattened wheat, barley, rye, and so on. Such formations, often extensive and highly intricate in design, have been found throughout the world in ever-increasing numbers since the 1970s. Much reported in the media, there was at first feverish speculation over their origin.
Among the favored theories were that:
Both explanations appear to fit the available evidence, so how do we decide which of these or the other available theories we should believe? In the absence of any other information, can we make a rational choice of one theory over its rivals? According to a principle known as Occam’s razor, we can: where two or more hypotheses are offered to explain a given phenomenon, it is reasonable to accept the simplest one—the one that makes fewest unsupported assumptions. Theory 1 assumes that UFOs exist, an assumption for which there is no clear supporting evidence. Theory 2 makes no assumptions about paranormal activity; indeed it assumes only the kind of prankish human behavior that has been common throughout history. So we are rationally justified—provisionally and always allowing that new evidence may become available—in believing that crop circles are the work of human hoaxers.
In fact, in this case Occam’s razor is spot on. It is now known that theory 2 is correct, because the hoaxers concerned have admitted as much. Is the razor always as reliable as this?
Occam’s razor is named after William of Occam, a 14th-century English philosopher. The “razor” comes from the idea of shaving off any unnecessary assumptions from a theory.
Ambitions and limitations Sometimes known as the principle of parsimony, Occam’s razor is in essence an injunction not to seek a more complicated explanation for something where a simpler one is available. If several alternative explanations are on offer, you should (other things being equal) favor the simplest.
Occam’s razor is sometimes criticized for not doing what it does not in fact set out to do. Empirical theories are always “underdetermined” by the data on which they are based (see Science and pseudoscience), so there are always several possible explanations for a given body of evidence. The principle does not claim that a simpler explanation is correct, merely that it is more likely to be true and so should be preferred until there are grounds for adopting a more elaborate alternative. It is essentially a rule of thumb or methodological injunction, especially valuable (one would suppose) in directing one’s efforts in the early stages of an investigation.
It is sometimes tempting for doctors, especially young doctors, to diagnose a rare and exotic condition where a commonplace and mundane explanation is much more likely. To counter this tendency, US medical students are sometimes warned: “when you hear hoofbeats, don’t expect to see a zebra”—most of the time, the more obvious diagnosis will be the correct diagnosis. However, as in similar applications of Occam’s razor, the simpler explanation isn’t necessarily the right one, and any doctor who only ever recognized horses would indeed be a horse doctor. US doctors working in Africa of course have to reverse their aphorisms.
The razor in action Although generally not explicitly credited, Occam’s razor is frequently wielded in scientific and other rational debates, including many that appear in this book.
The brain in a vat (see The brain in a vat) problem sets up two rival scenarios, both apparently compatible with the available evidence: we are real physical beings in a real world, or we are brains in vats. Is it rational to believe the former rather than the latter? Yes, according to Occam’s razor, because the former is much simpler: a single real world, rather than the virtual world created by the vat, plus the vat apparatus, evil scientists and so on. But here, as often elsewhere, the problem is shifted, not solved: for how do we tell which scenario is simpler? You could, for instance, insist that the number of physical objects is what matters and therefore that a virtual world is simpler than a real one.
Occam’s razor makes a rather unseemly appearance in engineering and other technical fields as the “KISS principle.” In developing computer programs, for instance, there is an apparently irresistible attraction toward complexity and over-specification, which manifests itself in a bewildering array of “bells and whistles” that are ingeniously bolted on and promptly ignored by 95 percent of end users. The gist of the principle whose application is intended to avoid such excesses is usually taken to be: “Keep It Simple, Stupid.”
In a similar vein, the other minds problem—(see Other minds) the problem of how we know that other people have minds—is sometimes dismissed with a flourish of the razor: all sorts of other explanations are possible, but it is rational to believe that people have minds like our own because attributing conscious thoughts to them is much the simplest explanation of their behavior. Again, though, the razor is seriously blunted by questions over what counts as simple.
The razor is often used against a range of dualist accounts, on the grounds that it is simpler not to introduce another layer of reality, level of explanation, and so on. Unnecessary complexity—positing separate mental and physical realms and then struggling to explain how they are connected—lies at the heart of many objections to Cartesian mind-body dualism. The razor may slice away one layer of reality, but it doesn’t of course indicate which one to throw away. Today physicalists—those who suppose that everything (including us) is ultimately open to physical explanation—form the great majority, but there will always be some like George Berkeley who take the other, idealist path (see The veil of perception).
Judicious use of Occam’s razor is supposed to facilitate rational choice between rival theories. Buridan’s ass—supposedly due to William of Occam’s pupil Jean Buridan—illustrates the danger of over-rationalizing choice. The ass in question, finding itself placed midway between two haystacks, can see no reason to favor one stack over the other and so does nothing and starves to death. The hapless beast’s mistake is to suppose that there being no reason to do one thing rather than another makes it irrational to choose and hence rational to do nothing. In fact, of course, it is rational to do something, even if that something cannot be determined by rational choice.
A blunt razor? The idea of simplicity can be interpreted in different ways. Is the injunction against introducing unwarranted entities or unwarranted hypotheses? These are very different things: keeping the number and complexity of hypotheses to a minimum is sometimes referred to as “elegance”; minimizing the number and complexity of entities as “parsimony.” And they can run counter to each other: introducing an otherwise unknown entity, such as a planet or a subatomic particle, might allow a great deal of theoretical scaffolding to be dismantled. But if there is such a basic uncertainty about the meaning of the razor, is it reasonable to expect any firm guidance from it?
the condensed idea
Keep it simple
| Timeline | |
|---|---|
| c.350BC | Forms of argument |
| c.AD1300 | Occam’s razor |
| 1637 | The mind-body problem |
| 1739 | Science and pseudoscience |
| 1912 | Other minds |
| 1962 | Paradigm shifts |
| 1981 | The brain in a vat |