-The explanation requiring the fewest assumptions is the most likely to be correct.
-Whenever two hypotheses cover the facts, use the simpler of the two.
-Cut the crap.
~ Alternative statements of Occam's Razor
Back in the late 13th or early 14th century, one William of Occam (or Ockham as it was spelled then), a Franciscan monk and philosopher, opined, “Pluralitas non est ponenda sine neccesitate.” To those of you whose Latin stops at “Et tu, Brute”, William said, “Plurality should not be posited without necessity.” The phrase has become known as “Occam's Razor” and has been restated in many different ways, including those above. Note that none of those statements (except possibly the third) really reflect what William was saying very well. There is a very good explanation of how Occam's Razor has been used and misused over the years here. The point, I think that is being made, is that a) William's statement has been modified in use over the years, and b) it's used by people on both sides of many arguments to justify their approach.
I prefer an interpretation best stated here, which translates William's Latin as “ one should not increase, beyond what is necessary, the number of entities required to explain anything.” In other words, once you have a hypothesis that explains the observations, you should not go gilding the lily. Consider that you have an observed phenomenon, which can be explained by hypothesis A or hypothesis B. A provides a full explanation; B also provides a full explanation but requires several additional elements be present to be valid. Or, B may require that other well-established theories be reworked so that B will be valid. In either case, hypothesis B needs to be viewed with a grain of salt.
It's not that speculative hypotheses are a priori bad; quantum mechanics is a vast playground of out-there theories, all trying to explain what's going on down there in quarkville. But, that understandable because there's so much we haven't been able to determine at quantum levels. Similarly, when speculating over cosmological questions, we have to expect the scientific mind to go way out there because our observational information is constantly increasing, and what we observe keeps changing our outlook. In other words, sometimes you have to guess at what's going on because you don't have enough information yet.
That being said, some theorizing often seems to be speculation for the sake of publication. Here are a couple of recent examples.
As Cassini has been chugging around Saturn, it has revealed all manner of wonders among the moons of the planet. For example, Enceladus apparently has geysers or “plumes” like those seen on Triton by Voyager. The going explanation for this phenomenon has been tidal warming. Saturn's gravity “massages” Enceladus much the way Jupiter works over Io. On Io, that results in volcanism; on Enceladus, it's geysers. But another group of researchers has what they think is a better idea: Cosmic rays.
Now cosmic rays are everybody's favorite phantom. It's not that they don't exist; they are most certainly real. But, they are credited with many effects from causing mutations (which they probably do) to causing computational errors in computers (which is much more unlikely). The team in Maryland thinks that cosmic rays react with water ice on the surface of the moon, breaking loose free oxygen, which then seeps into the planet where it meets with ammonia ice. Oxygen and ammonia react violently when brought into contact, which would cause heat that would result in water being violently rushed to the surface.
Very elegant, except for one thing: Cassini has not found ammonia on Enceladus. Moreover, as another scientist points out, Mimas, farther away and smaller than Enceladus should have geysers also, if cosmic rays are the cause, but it doesn't.
In other words, tidal heating explains everything we've so far observed about Enceladus without conflicting with any other information we have. The cosmic ray theory needs something we haven't seen (ammonia) and requires further amplification to deal with places like Mimas. Occam's Razor comes down in favor of tidal heating.
On a grander scale, a group of Italian scientists think that the universe is shaped like a pill, not a sphere. The basis for this new view of the cosmos is none other than our old friend, the cosmic microwave background (CMB), which has been a topic in this blog recently on a couple of occasions, most recently here. Looking at data from the Wilkinson Microwave Anisotropy Probe (WMAP), they have found that, when sampling large areas of the sky, microwave radiation seems to be under-represented, while on small samplings the radiation is at expected levels. Their explanation for this is that the universe is ellipsoidal, not spherical.
Now, we're not talking about a universe shaped like a football; the eccentricity is about 1%, but at cosmic distances, that's meaningful. So their theory explains an observed variation; what's wrong with that?
What's wrong is that it doesn't begin to explain how the universe would have become ellipsoidal to begin with. There's no mechanism in the Big Bang theory, which is otherwise well-supported by WMAP, that will cause this eccentricity. As one scientist put it, to explain a “mild anomaly”, the Italians have introduced an unexplained new feature of the universe. It's quite correct to point out anomalous data; it's yet another thing to change the shape of the entire universe based on that single data set.
The article supplies a hint for the anomaly at its end when it mentions that the theory will be tested when future CMB-measurements are taken by satellites that can analyze the polarity of the microwave radiation. In other words, variations in polarity, rather than squishing the universe, could explain the data.
Once the data is in, we'll see if Occam's Razor slices and dices or not.