09 April 2006

When a science isn't one and why it matters


These are the rankings of the hard sciences as seen by scientists themselves. Mathematicians look down on physicists because physicists take the wonderful realm of ideal number systems and actually *use* them for something. They pull down sweet mathematical castles to use the *bricks* to make something else, the barbarians. Physicists tell you all about the structure of the universe and atoms and how they work, and, in turn, look down at chemists because they actually *use* those theories to do things. Such ingrates. Needless to say the chemists look down on biologists, because all of that wonderful chemistry is put to use doing things in biological systems. How dare they make *use* of sweet chemicals in search of actually doing something in an organism? And the biologists look down on geologists as, well... no one ever really explained that well. Geologists just try and make it *all* fit together in a reasonable way. So putting everything to use gets you pretty much looked down upon as the worst of the worst.

However, all of these sciences have one thing in common: they make hypotheses which can be tested and either have the hypothesis validated or invalidated. Once sufficient validation is present a hypothesis becomes a *theory* and is used as a working principle to explain things. A theory makes *predictions* based on how the theory works and changes things by its internal components, usually math based. Any system claiming to be a science must work towards this goal: taking observations, formulating a hypothesis, testing it, then either validating or invalidating it and applying such to current theories and then looking to see if something in current theories needs adjustment.

Starting from four elements and Plato we are now at post-Einsteinian physics and working to integrate as much of the physical world as possible, taking in Quantum Mechanics and multi-dimensional worlds or other forms of twisted space-time. The goal is to give a good description of how the Universe works and ensure that good and accurate predictions can be made given a set of circumstances. Chaos theory will play its role, as will the Quantum scale, but in the whole a good and reliable way of working with the Universe is the goal of the hard sciences.

The social sciences, meanwhile, have added mercury to the four elements, but really don't want to deal with it, even though it is there.

What do I mean by that? Well lets take a science-human interface and take a look at it. I like earthquakes, for the reason that we have good understanding of the forces behind them, but the ability to predict where and when they will happen *exactly* is difficult if not impossible. Now many folks will point to *animal* reactions that might imply an earthquake. First, this observation is usually made ex-post facto or after the fact. For this to be a useful indicator the animal or animals must show a high degree of correlation between unusual activity of specified sorts and earthquakes. If the animal or animals show such activities during times that are *not* followed by earthquakes, then a simple look at what the percentage degree of correlation actually *is* would be helpful. Also a look at animal activity and degree of earthquake activity would be *extremely* helpful: do the animal(s) act in such manner for a magnitude 4 and a magnitude 7? Or is there a climbing degree of activity?

So, if animals only act partially in correlation to earthquakes and will often display said activity during non-earthquake events, then there must be a higher than statistical average correlation for this to be a helpful *indicator*. Peg that at around 20% on the utility scale. Anything below that falls into the happenstance and chance region, and anything above that shows a good degree of correlation. That said anything below the 90% range makes the correlation just one of *many* that needs be considered and then weighting given to it and other indicators.

Now, lets run the experiment of this indicator having high degree of correlation and being used as one of many factors, but not a leading factor in earthquake prediction. Small zoos are set up along earthquake prone regions to test the validity of animal types, mix and behavior. And at some point a decade or so later it is actually used to help predict earthquakes. And an earthquake is predicted.... and doesn't arrive. The correlations show up *again* and the prediction is made... and the earthquake doesn't happen. Now what will be the response in the same area to a *third* such prediction?

A metropolitan area subject to such a thing will do precisely *what*? And there can be no hand waving, no implications, no recourse to saying 'well human nature...' Just what, within reasonable error bars, will be the response of a metropolitan area to a third warning when two have proven false?

So let me take a look at those things which are *closer* to being sciences, but are not there yet:

Psychiatry - Measuring how medicines impact an individual's psyche and change things is very difficult. The first being done, today, is to find out exactly why some medications will have sudden and unforeseen mental side effects and if those are caused by the nature of the medication or some underlying structure of thought and how it interacts with that chemistry. Yes, Human Nature, again. By having to have one foot in the biology area and another in the psychology area, psychiatry will be pushing the hardest of all to move into the scientific realm and, possibly, categorize interactions with medications as part of biology. Add in the Neurosciences (studying the behavior of brain activity) to this and psychiatry may see breakthroughs into the descriptive and predictive realm of science within a decade.

Economics - Often slandered and I will be the first to own up to that. But with good reason, as the basic economic proposals are, at best, hand waving at how societies actually *work* economically. Humans, being non-linear thinkers, behave in non-easy to understand modes of thought. Even the best of models in economics fall apart under real world activities within a day or so. And putting non-linear software systems in place does not help this fact as the changes done by the software itself are non-linear and difficult to figure out for evolutionary algorithms. The basic principles of mathematics help out much in economics, but by not understanding reactions to events and how they sway individuals and entire markets, economics is far from being a descriptive and predictive science. Again, not understanding Human Nature hinders this greatly.

Anthropology - These poor folks are the hardest drinkers of all the realms in the sciences and near-sciences. Geologists throw cases of beer into nearby rivers and pop a cool one when they get back at camp to figure out just what it is they saw. By working in harsher climates and needing something a bit stiffer, anthropologists make geologists look like teetotalers That said, they are doing the hard work of understanding the rise and fall of societies due to climate, geography, diet, religion, and, yes, human nature. A very descriptive capability that is laying the foundations for much future work, but is again hampered by our lack of understanding of individuals and how they operate in groups as a whole.

Marketers - What? This is a near-science? Well, look at it: it addresses segmented populations which it defines to try and get verifiable results. That is almost a classical description of science! It is descriptive, predictive and uses feedback to adjust its viewpoint. There are a lot of marketing 'misses' from your daily Spam to New Coke, but, by and large, they have done the best of anyone of being able to define groups and market niches and then find messages attractive to those niches. They push everything from political messages to, sadly, New Coke. And as that last shows, they still do not understand the foundations of Human Nature. This should, in actuality, BE taught as its own subject in College and Universities and, perhaps have its own department as a science! Not a social science, but as a science.

Warfare - The Art of War is a common phrase and it still is an *art*. But from Sun Tzu to Machiavelli to B. H. Liddell Hart there are the basics of understanding that are becoming synthesized within the modern military realm. The ability to take in enemy activity, adjust ones inputs across a wide scale of options to stop an enemy from enjoying success or even taking the field is an art and becoming a science. Today's military training now emphasizes social structure, local languages, interacting with the local culture, how to win friends, examining opponent hierarchies, leveraging combined arms assault to do the most harm to one's enemy while limiting other damage, construction, field work... the list is literally endless. Consider this anthropology with a pointy end to it. Being able to flex across a wide variety of networks from the social to the tactical to the Theater to the Strategic is something that is now being pushed downwards into the lower ranks of the US Armed Forces. By putting in place a harsh and constant feedback loop from the TRAINING end, the result is a highly adaptable and flexible organization that has a root understanding of how the human condition works under stress. What it does NOT have is a way of easily describing those factors, describing interactions and quantifying them. Thus it lacks the hard, numbers based descriptive capability which is a necessary theoretical basis of a science.

Psychology - The descriptive end of human mentation at a personal level. The main problem with psychology is that it does not qualify nor quantify enough to understand the exact measure of a person's mind. While the mind is a squishy thing, and has vagaries due to its chemistry and structure, the concept should be understood that thought *structure* is a good description for those underlying factors. But even the basics of temperament, psyche and intelligence are only vaguely qualified and making *predictive* assessments is barely an art and more a collection of 'Rules of Thumb'. There is much that could be done here to help the entire set of works looking at Human Nature, but it is hard work to do so and will require long, long studies before results are seen. This is a KEY area for unlocking human potential and understanding of how people work, but is still at the Witch Doctor stage, although Psychologists are pushing that hard, too.

Painful isn't it? Taken as a group we have tended to elevate one or more of these at a time looking for insights into the human condition and each has had its glory days in the sun followed by the setting as Human Nature slaps it upside the head from an unexpected angle. As a whole, they could come together to form a synthesis for understanding how people work at all scales, from the individual to the National. I will get to the hard work topics in a bit to do so, but next up are those 'sciences' which are not.

These folks are not up past the hypothesis stage and have no good way of taking in feedback to understand if their very hypotheses work or not:

Political Science
Ethnic Studies (all of them)
Gender Studies (the non-biology portion)

Not a single one on that list is even as far as Marketing and they could ALL use the rigor of that as an example. These are NOT sciences. Understand that this is not to say that they are good or bad, but is a representation of their lack of scientific rigor in inquiry and thought.

And how do I propose that such rigor gets added? Simple, really. Define Human Nature with mathematical backing and understanding. Any and all hypotheses must use those definitions, the math that goes with them and make descriptive and predictive postulations that can be *tested* and are verifiable via analysis and direct comparison with human beings.

Now how does one go about defining Human Nature with rigor? Well there are a large number of personality tests out there, and one of the more descriptive and thorough sets are the Myers-Briggs Type Indicator for personality. As in any science a measurement capability is needed so as to properly define what is being examined. A personality test is such a measurement tool and the better the tool is able to define personality the more useful it *should* be. Yes, the test has some problems, but by not having postulational rigor behind it the improvement is haphazard at best. But as a *descriptive* tool it has utility and can serve as a basis for understanding human temperament.

Human intelligence is a multifactor area that is not well addressed anywhere. The concept of an Intelligence Quotient only measures for *some* things and not others. How can something like the ability to interact with groups of people well be measured as a factor of Intelligence? How about something like 'street smarts', where does that fit in on Intelligence? Basically there are a whole raft of types of Intelligence that are not well defined and the entire area needs qualification and quantification: definition and measurement.

Then there are the groupings and societal concepts that move segments or entire populations. Politics is one area that is extremely ill-defined and requires much rigor just on definition terms and underpinnings. What is the measurement of something like Liberty? Or of Equality? Or of Justice? And then how do those play out as defining terms for a group or Nation? And just what are the impacts on Intelligence and temperament within that and how do those change the larger scale measurements? And I am *not* talking about hand-waving and giving postulates. I mean real terms that can be numerically evaluated, applied via formulas, tested against populations and then have their validity checked.

That entire concept of actually having to define something and then give its mathematical relationship to other things within that system is one that helps put rigor into the scientific method for the 'hard sciences'. Giving vague postulations and assertions is not only unhelpful, but it represents stasis and some decay of the underlying system doing the examination. By not actually having the courage to put definitions onto human thought, numerical valuations to those definitions and mathematical descriptions of how those valuations interact with each other and what it is they describe, the entire field of the 'soft sciences' does a disservice to mankind by putting out postulations for things that may only be valid in limited circumstances, if at all. Further, by making them non-testable or only testable in vague ways, it gives a patina of validation to something that is only apparently good.

Until that point in time, I shall remain a Jacksonian and measure ACTIONS and what their impact is. By having no clearly defined and validated way of measuring *intent*, the application of labels via supposed intentions is one that is subjective and used for retribution and silencing people and their ideas. Put up a good and internally sound description of what human intent is based upon, how it interacts with personal temperament and then changes thoughts so that actions result, then I may listen to you. And check your numbers on my own, as any good scientist would.

Liberals and Conservatives *both* refuse to properly define what they are purporting, demonstrate their interaction on the sliding scale from individuals to Nation States and then put forth mathematical rigor on how their positions actually CHANGE society and human nature. As neither can do that, I am left to personal wisdom, insight and the conceptions put forth in the Constitution. Give the Individual all the rights save those few necessary to *govern* and let them figure out what is the best way to apply those rights. These things have not been clearly defined in a numerical fashion to allow for the checking and cross-checking of the principles put forth by Liberals and Conservatives. They can NOT do this as there is no *science* to allow this to be done.

That said the formulations used by Jacksonians are simple and straightforward. This is the social contract that We the People have agreed upon. And since no one has come up with anything that is mathematically proven to be *better* these stand as time tested concepts for our society. We ignore them at Our peril as the results are not mathematically known, but other societies that have tested these in a negative way have quickly deteriorated:

1) Agree upon what the laws are and their penalties, then enforce them!

2) For those that give honor or have shown that they are honorable in action, give honor.

3) For those that give friendship, grip their hand and hold them and give friendship in return.

4) Let others live and let live as they will so long as they do not go about breaking the law or the compact of the Constitution nor try to enforce their strange notions upon you.

5) To those that seek to harm you or your Nation: Hound them, deny them any succor or rest, and if they have acted dishonorably and seek your end, then kill them.

So, until all those highly learned folk actually try to put some rigor into their work, they can keep their petty labels to themselves. I label actions, not intent, as no one has actually figured out how to DEFINE AND MEASURE INTENT. And until that point in time, where it is defined, validated, tested and can be shown how intent is an outgrowth of temperament, intelligence, national character and belief systems, I will continue to spurn those painting with the tar brush as it paints broadly and not well. And those doing the painting actually ARE the problem now, because their actions are causing harm to the underpinnings of the Constitution and rule of law.

They may claim to be doing something *good*.

But they have ZERO to back them up on that.

And wanting to stick with what works and is not broken is *not* conservative.

It *is* smart, however. Unfortunately some are too intelligent to see that.

No comments: