Theories

What is a theory? At first this seems like a rather unnecessary question. However, it is not. Unfortunately "theory" is a word that exists equally in the realm of common discourse and in the realm of professional science, and it does not automatically mean the same thing in both contexts. This has been commented on abundantly in discussions of of evolution "theory", such that the common imprecise use of the word is often conflated with the technical use in attempts to discredit the concept of natural selection. The archetypal misuse of the word typically appears as "such-and-such (e.g. evolution, or the Big Bang) is "only" a theory). There may not be a definitive set of meanings for the word "theory", and I am certainly not an etymologist. From here on I write idiosyncratically, that is according to my own best understanding of the word.

In my understanding, the word "theory" as used in common speech is more or less equivalent to the word "idea". Thus one could say, I have a theory as to why men have more body hair than women, or, I have a theory as to why the store across from the gas station closed last month, or even, I have a theory that the moon is made of cheese. In this sense, essentially any statement of explanation can be called a theory. One would not normally say, I have a theory that this food is too salty, since that is simply a restatement of an observation. One might say however, I have a theory that this food is too salty because the cook put in too much salt, although such a statement appears rather self-evident. Thus even in common usage the word theory carries an implication of some kind of explanation going beyond a pure observation.

However, in significant distinction from common usage, a theory to a professional scientist is not simply any such statement of causal explanation. A professional astronomer would not be likely to theorize that the moon is made of cheese to explain why its surface looks holey, or that stars are caused by small holes in a dark globe that surrounds the earth outside of which is a universal source of light. Nor would most marine scientists theorize that the tides are caused by the gods blowing on the ocean with their breath. To a scientist, a theory is indeed a statement or more often a set of statements. However, these statements must have certain properties: First and foremost, they must possess explanatory power for a set of direct experimental observations. Thus, the observation of tides on our world's oceans can be explained by assuming that the moon exerts a gravitational force on earth, to which the water in the oceans responds in a measurable way. Moreover, by various mathematical operations one can deduce a specific number for the likely mass of the moon, which together with its measured size, can be used to calculate a density which is much greater than that of cheese. Such calculations were performed long before astronauts actually walked on the moon, observing at first hand that it was indeed made of rocks and minerals as predicted. The more observations that are explained, the stronger the theory. Thus the theory of gravity explains not just the tides, but the orbits of planets and their elliptical paths, as well as many other astronomical measurements.

This example also points to the second major component of scientific theories, namely that they posses predictive power. The theory of gravity, together with specific measurements made in laboratories on earth, allows one to make predictions about the behavior of other systems (such as the oceans and the moon) for which such direct measurements (such as weighing the moon) cannot be made. Moreover, these predictions must involve some fairly specific obligations, namely observations of a system made under circumstances different than those used to develop the theory. These can involve either an experimental perturbation of the system (admittedly hard to do with the moon and oceans), or observation of the system with a new technology or under some new conditions. This in fact could be done for tides, given that the moon's distance from the earth is demonstrably changing due to the extremely slow decay in its orbit. Presumably such changes in the moon's orbit will eventually lead to measurable differences in oceanic tides, which could be measured if one waited long enough. Alternatively, if there were a way to measure the properties of tides many millions of years ago, when the moon was at a different distance from the earth than it is today, this would also constitute a type of historical prediction. In fact, the observation of the changing number of days in the year, documenting the loss of energy from the earth/moon system and the concomitant decay of the moon's orbit, as determined by counting rings in ancient fossil trees, it itself a major successful prediction of gravity theory that Newton probably would have greatly appreciated though he might not have ever thought of it.

In a very important sense, scientific theories therefore are actually simplifications. If one has a large set of observations, such as the height of sea level at various times of the day and year at various seashore locations, a theory of tides must somehow reduce these observations to a simpler set of statements. If one has a large set of observations of the luminosities and distances and speeds of various astronomical objects (starts, galaxies, quasars, and so forth), a theory of the origin of the universe must account for these quantitatively, and in principle make predictions of the future of the universe that could be tested over time.

In fact, there is a word in scientific usage which to my mind is more in keeping with the common usage of the word theory, and that is "hypothesis". To me as a scientist, a hypothesis (an hypothesis?) is a statement that may explain one or a small set of observations, but without having necessarily put large amounts of effort into testing any predictions of the hypothesis, or even being sure whether it can make predictions. The distinction is subtle, but very important. The theory of evolution by variation and natural selection entails a set of multiple statements, all very sophisticated, potentially explaining a multitude of observations about the diversity of life on earth, the physical appearance and behaviors of all plant and animal life, the basic biochemical basis of life, the extreme similiarities of DNa sequences in the genomes of organisms, and so forth. Thus it constitutes a scientific theory. On the other hand, statements such as "men have more body hair than women because they have more muscle mass and sweat more" or "men have more body hair because women happened to like the look so such men had more children" are hypotheses, as they are at least at first glance efforts to explain a very limited set of observations, and in fact may not yield any testable predictions.

From this perspective, it is quite unusual for scientists to express major disagreements about the status of current theories. That is simply because almost by definition theories do a good job of explaining the available observations. To the extent that a current theory fails to explain important results (such as the equivalent speed of light in various geometric configurations in the Michelson-Morley interferometry experiment) in fact argues that the theory is incorrect or at least invalid for some particular circumstances. Eventually, in the face of many discordant observations, theories are at last updated, replaced or simply nullified. This last is rare however, since the theory arose to explain many previously made observations. More often in science, theories are shown to be oversimplifications, valid only under certain circumstances. The Newtonian theory of gravity, mass and motion is just as valid today as when Newton theorized it. However, it is an approximation that become increasingly inaccurate at higher speeds (approaching light speed) and for very high mass objects (such as stars), circumstances for which Newton had no observations to confound his equations. The theory of relativity is in that sense an update, or complexification of the older theory, with more sophisticated mathematics and usable under a much wider array of circumstances. As long as one is just looking at the planets far from the sun however, Newton's equations of motion remain useful.

In consequence, scientists place much more weigh and respect on theories than hypotheses. Thus, although as critics of evolution are quick to note, biologists disagree on many aspects of evolution theory, in fact there is no serious disagreement among professional biologists on the theory itself, meaning its major statements as beautifully summarized by Ernst Mayr in his many essays, or by Douglas Futuyma in his classic textbook. We may disagree on whether new species always arise in geographic isolation or only sometimes, or how often. We may disagree on whether a particular bone's structure proves that birds are descendants of dinosaurs (generally thought to be true) or whether the diminutive fossils of Flores island in Indonesia result from a truly different species of human or from normal genetic variation within the species Homo sapiens (not universally agreed as yet). But such arguments do not imply any doubt of evolutionary mechanisms per se. Evolution by natural selection is a scientific theory like any other, and could indeed be "proved" wrong or more likely shown to be oversimplified. At this time professional biologists would be hard put to imagine what kinds of data would disprove the theory. However, short of accepting the verbatim validity of the Hebrew Old Testament (why just the OT, and not the Vedas or other religious texts is not addressed by critics), there are no generally accepted biological observations inconsistent with evolution theory. Even the recent and fairly sophisticated argument that the current theory of life arising as an RNA molecule is impossible since RNA cannot be made from small organic molecules in the laboratory, has recently been invalidated by some wonderful biochemistry demonstrating the organic synthesis of RNA-like precursors from simpler molecules.

The nature of information, and the definition of simple versus complex, is surprisingly tricky. Scientific definitions, while in a very general way similar to common usage of these words, differ in full technical detail, so that one's intuition must be trusted with extreme caution. For example, those gorgeous fractal patterns one can see everywhere nowadays, are generated by very simple mathematical formulas. Similarly, a pretty simple mathematical algorithm, when run for a long time, generates the value of pi to as many decimal places as one could want. Pi is an irrational, and moreover a transcendental number, which never repeats and has no obvious internal "pattern", yet can be generated by a simple calculation. A truly random series of numbers has no internal pattern, and by definition therefore cannot be generated by any simpler formula or calculation, it can only be described by writing the number itself to as many digits as there are. But for a given large number, with no obvious internal pattern, it's very hard to be sure that it cannot in fact be generated by a simple algorithm (i.e. set of calculations). If I understand correctly, it can be proved that it is impossible to prove that there is no possible algorithm to generate a given output. So, short of actually finding a simple algorithm to generate a number (like pi), it's hard to know that a given number is truly truly random - maybe someone smart enough will figure out the algorithm in the future.

So much for pure math. The human genome has approximately 1.5 billion "unique" bits of information (i.e. GATCs), which can be thought of for some purposes as a code or "number" containing data, but it's very hard to known how much actual information is contained. By comparing our genome to those of other animals, some estimates are that only about 1% of our unique DNA sequence is highly conserved, though other estimates are much higher - it depends to some extent on which other animals are used for the comparison. At this point in time we don't know the answer.

So much for the genome in the abstract. There's no question regarding the tremendous complexity of the human brain in terms of the absolute number of neurons (cells) and synapses (links). But it's very unclear how much genomic complexity is required to create the structural complexity. Maybe the underlying algorithm is very simple, and is simply repeated a large number of times, like fractals, only with lots of feedback from the environment so that no two human brains end up working the same way. I like to keep in mind that many types of plants are biochemically much more "complex" than we are, in that they can carry out many more types of chemical reactions and have more different types of small molecules (like glucose, ATP, acetyl-CoA, etc) than we do. It is somewhat animal-o-centric to hold that complex 3-dimension bodies with lots of different tissue types are "more complex" than complex biochemical networks. Plants might feel that they are actually more advanced, and better adapted than we are too. If the amount of oxygen in the atmosphere dropped from 21% to say 15%, I imagine most living animals, or at least mammals, would quickly go extinct, whereas many plants could probably adapt quite well. Superior adaptation to a specific environment does not necessarily equate with overall success under environmental change. Let's see who's left standing 100-200 years from now, after the global average temperature goes up 5 degrees C. I predict there will be a lot fewer humans, but probably about the same number of cats!

Well, as the meanness extends fully into the Democratic party, it seems to be more than ever the case that "they" in power are "all the same". How depressing. If both parties have been fully captured by oligarchic interests (presuming the Dems in the past were only partly so), then indeed what is left to do? Voting for a celebrity nut like the D is no answer. Voting for Bernie even as a write-in is at least sending a message, although it's hard to see what practical effect it can have at least in the short run. 100 years from now perhaps they will look back on this election as the last gasp of truly unfettered dog-eat-dog capitalism, not even a pretense that the government's job is to protect the well-being of the majority, which led to the (yet to be seen) american social revolution. I don't expect to be around to see such a future, and if it comes I suspect it will entail a fair amount of violence.