Philosophy - Ethics - Scientific Prescription

The very beginning of ethics must have been a prescriptive beginning for the prescriptive variety to come be known today as “normative”, a word that is used in many fields to describe the typical or expected.  There certainly could not have been meta-ethics, as that discipline requires a number of established ethical systems to compare and contrast.  And descriptive ethics seems to be a purely modern “science”, concerned with pointing out to would-be prescriptive ethicists the anthropological limits of what they are allowed to propound, denounce, prohibit, and exhort, if they wish posterity to view them with anything but flabbergasted curiosity.  Where the prescriptive ethicist is concerned with possibility, potentiality, and the experimental probing of conventional habit, the descriptive ethicist is content to observe passively and take note of what has been the case, then report back to the rest of humanity with their data, in much the same way that a physicist might matter-of-factly record the natural laws of gravity, insofar as they have observed them.  So the question that leaps to my mind is, which of the two is most truly deserving of the adjectival accolade “scientific”, which the descriptive approach has seemingly won for itself?  
Let’s investigate.  

Every branch of science is characterized by a methodology consisting of four essential steps – characterization, hypothesis, prediction, and experiment.  That is, some aspect of reality is observed, a hypothesis or tentative guess is offered to explain it’s origin and exact nature, a prediction is made about what the upcoming experiment will reveal, then the experiment is conducted to test both the aspect in question and the accuracy of the experimenter’s intuitive understanding of that aspect.  For example, the famous renaissance physicist Galileo Galilei rolled balls of dissimilar masses down an inclined plane to test whether an object’s mass affected it’s gravitational acceleration.  The paradigm-smashing proof he procured with his experiment was that it did not.  

Now let’s examine Lawrence Kohlberg’s theory of moral development.  Kohlberg developed his descriptive theory by asking children morally charged questions – such as whether or not an expensive drug should be stolen by a man to heal his terminally ill wife.  These responses constitute the characterization – the observation of the discriminating human moral sense in action.  Kohlberg then divides the reasoning the children offered for their responses into six categories within three tiers.  This may constitute a hypothesis, though we are not graced with the reasoning behind it’s seemingly arbitrary formation, if there is any.  This is a theory of moral development; are the ages of the participants taken into account, and if so, are the responses of the older taken to be the more developed?  Is there a consistent display of age-to-justification uniformity?  Egoism rests at the bottom of the pack (implicitively least principled in comparison to the top-tier “principle-driven” stage), common among children.  The deontology of Kant enjoys a lofty perch in stage six individuals, who have apparently transcended the pernicious influences of authority, peer pressure, and personal greed.  An eminently mortal hypothesis, but an hypothesis.  Now what?  Where is the prediction?  More importantly, where is the experiment? 

An experiment would attempt to measure what it is possible for humans to view as norms and taboos.  Walking into a “no shoes, no shirt, no service” restaurant without any pants on would be an experiment, where the control group is the restaurant prior to your entry, the experimental group, the restaurant after your entry.  Raising a Spartan child used to stealing at whim, then attempting to reason to them at their maturity the immorality of theft would be an experiment, where the control group is the child acting in accord with habit and custom before your reasoning, the experimental group, the child after your rational and reflective sermon to them.  That these groups are temporally rather than spatially distinct is of tangential importance.  Most pertinently, these experimental and thus fully scientific ethics demand and imply prescription, revealing the descriptive approach as the mere semi-science that it is.  An ethicist that is not attempting to change human behavior cannot be attempting to understand it, for to change it is to understand it.  In much the same vein of logic, a physicist that is not attempting to change “gravity” (as the acceleration-dependent-upon-mass phenomenon that it is) cannot be attempting to understand it.