, , ,

Within the limits of my own study, the idea of “data density” and its relation to science vs. pseudo-science, is not one that I recall having encountered. (And I freely admit that my studies are limited; the world is large, and human life is short.) I suspect that no small part of the problem is that we have only begun to slam into this wall in earnest in the last few generations. I wish to use this idea of “data density” here to compare two branches of scientific study. It is my thesis that the data in gravitational cosmology is especially “thin” and “patchy,” which makes the general lack of attention to alternative models to the standard one especially inexcusable.

Thin grass

Thus, for example, Alfred North Whitehead’s alternative to Einstein’s general theory of relativity was proposed in detail in 1922. Yet, even though it fit the evidence every bit as well as Einstein’s theory (and in a couple of cases, better) it was all but entirely ignored, with no effort invested in it by theoreticians. It needs to be well and carefully noted that, by the time it was supposedly “refuted” by observation, that refutation could only pretend to stand by ignoring the fact that Einstein’s theory had been endlessly modified and parameterized so that it would fit the data, an advantage that Whitehead’s theory never enjoyed. It should also be noted (although, with only a handful of exceptions, it never has been) that Whitehead did not propose “a” theory, he proposed an entire family of them; but again, no effort was ever expended in fitting Whitehead’s theory to the relatively few independent observations available that was invested in guaranteeing that Einstein’s general relativity “worked.”

I am coming to suspect that no small part of the reason for this blatant, ideological disregard for viable alternatives was and is driven in no small part by the shocking lack meaningful evidence. The evidence, the observational data, in gravitational cosmology (the playground of theories of relativity, whomever might be their originators) is disturbingly “thin.” In comparison to the number of ways you can “tweak” your model to fit the data, the number of genuinely independent observations are almost vanishingly few.

Consider this analogy: suppose you watch a single episode of a television series that you’ve never seen before. Perhaps it is a pretty good episode, but can you really use that single episode as a definitive guide to the skills of the writer/director? Suppose you watch the episode again, this time varying the sound or the picture quality a bit at your television set. Do you really have any new information about the writer/director? Can you really make any definitive statements about the character arcs of the individuals portrayed in the episode? Of course not! For that, you would need multiple lines of independent evidence (numerous episodes, other examples of the writer/director’s work) in order to formulate an intelligent sense of the writer/director’s style and skill, and to gain some sense of how characters in the the show were developing over a season or over the entire series itself (assuming they do have an arc, and are developing.) So, for example, speaking for myself: having watched and rewatched most everything that Joss Whedon has written &/or directed – many, many times, in fact – I would be prepared to pay money to watch my calculus instructor from college reading from the phone book, provided Joss Whedon was directing it. (In fairness to that instructor, even mathematicians find intro calculus boring to the point of utter stupefaction.) I have multiple, independent lines of evidence regarding Whedon’s qualities as a writer/director (many of which include Whedon’s own presentations in the writer/director commentaries) such that my theory regarding Whedon as a writer/director is extremely robust.

Here is where the problems with gravitational cosmology (often called “Big Bang Cosmology;” for my part, I often refer to it as the “standard model of gravitational cosmology”) become acute.

As the physicist Michael Disney has observed, the number of independent observations (separate episodes in separate series) available in gravitational cosmology are significantly smaller than the number of ways (“free parameters”) in which the standard theory we use to model cosmology can be “tweaked” so that it never runs up against the embarrassing fact that it has been catastrophically falsified on any number of occasions. And here’s the thing that must be understood: every parameter that has been added to the standard model of gravitational cosmology was put there precisely because it was discovered that the model failed to fit observations. So many parameters have now been added that they outnumber the available independent observations, thus guaranteeing that it can be adjusted at whim to fit those observations with nearly arbitrary precision.

Let us compare this situation that which we find in climate science and the evidence pointing toward Anthropogenic Global Warming (“AGW.”) “Anthropogenic” identifies human activity as the principle source of forcing in global warming. Meanwhile, one would hope that by now everyone in the world would be annoyed at the redundancy of being reminded that there is a reason it is called GLOBAL warming, rather than “your back yard” warming. Sadly, that annoyance is not as universal, nor as founded in good science (some people get annoyed because they hate being reminded that they are spouting fatuous nonsense) as it ought to be.

The evidence for AGW is truly staggering in its density. It is not just that tree-ring series form an independent line of evidence from ice cores, but different sets of tree-ring series are independent of other tree-ring series; different ice core sequences are independent of other ice core sequences. Temperature series from different sources apply different statistical methods to extract their data: thus the HadCRU series (which is produced by the Hadley Climate Research Unit of East Anglia University) is independent from that produced by GISS (Goddarad Institute of Space Sciences) or NOAA (National Oceanographic and Atmospheric Administration.) The HadCRU estimates exclude measurements from the Arctic, because of the lack direct measurement, while GISS and NOAA incorporate what direct measures there are and apply various approximations and proxies; as a result, HadCRU consistently underestimates the degree of AGW because it does not take into account the massive amount of warming occurring in the Arctic. The examples above barely scratch the surface of the available evidence. There are many dozens, if not hundreds, of independent lines of evidence pointing unequivocally to the reality of AGW. When one couples this amazingly “thick” evidence with the repeatedly demonstrated fact of the overwhelming scientific consensus regarding AGW, then with nothing more than a basic appreciation for logic and rational thought, one can easily see that for all effective purposes, the case for AGW is scientifically made; denial, at this point, nothing more than the most loathsome ideological twaddle imaginable.

The situation is very different with respect to cosmology. The professional consensus regarding the standard model of gravitational cosmology is probably just as high amongst physicists, as is the basic understanding of AGW amongst climate scientists. But unlike the situation in climate science, the evidence supporting the tandard model of gravitational cosmology is so “thin” as to be almost spectral. As I have previously observed, we are presented with a patchwork chimera that is incapable of standing on its own feet.

Model-centrism is exactly this sort of valorization of the model over the data. One can identify sociological reasons for model-centrism, but at the end the day it cannot be legitimized on anything like logical or scientific grounds. Model-centrism is decidedly anti-scientific – which is ironic, since it is generally people who are considered leading scientists who are implicitly pushing for it – and it can ultimately breed the faux “justifications” for overt pseudo-science.