History is the dataset for which political theory generates hypotheses.
Political science generates testable hypotheses, which give more reliable results about matters of limited scope.
Historical data ad nauseum
Limitations of Experimental Science
History by hypothesis
This leaves the interpretation of experimental results dependent upon more encompassing conceptual frameworks — theories or paradigms — replete with axioms and other assumptions not directly testable.
Yet the predictive value of such experimental results can remain astonishing high and, thus, useful.
For example, though Newton’s theory of gravity was always threadbare (it bypassed how gravity worked) and has been thoroughly eclipsed by Einstein’s theory of gravity, it is still used for moon shots because the calculations required are much less time-consuming and hence costly than those required in Einstein’s theory.
Only for interstellar launches do the cumulative errors of Newtonian gravitational calculations become significant in practice.
The Gap between Science and Practice
But the big questions provide the encompassing interpretive framework for understanding the data but are not tightly determined by it.
They function more like axioms that lend coherence to the conceptual framework within which the data are interpreted.
They also provide the overarching ends that practical policies, however data-driven, are meant to serve.
For data are just observable remnants of causal processes, and causal processes operate without reference to ends.
That’s what makes them unbiased. They proceed from the perspective of universal consensus imposed by the scientific method on anyone operating within its remit.
For the subjectivity of interpretation and theorizing outside the severe limits of the scientific method comes from observer perspectives biased by the observer’s prior convictions, by their preference for hypotheses originated, championed or adopted by themselves, or by the observer’s values and beliefs.
Agency, Perspective & Practice
Only agency of some kind can supply ends — whether this is the survival instinct of a worm or the lofty deliberations of a statesman.
For only agents — or at least sentient subjects — have the power and need to preserve their internal constitution, the thing that makes them an individual organism with a unique perspective awash in threats and opportunities.
And without those, the concept of ends has no meaning, for causation works toward no specific future end, being wholly determined by the forces imparted to it in its causal past, not its future prospects.
To claim otherwise is to think about ends in isolation from the only context that defines their function. It is, the mother of all category mistakes.
In other words, the universal laws which guide causation have no concern for particular outcomes. Where natural selection is careless of individuals, individuals don’t even register with physics or chemistry.
In sum, only individuals have ends, need them, or are equipped to work toward them.
And to interpret and evaluate historical and political ends, empirical political science is a crucial stepping-stone, but political theory is the capstone.
When All Is Said and Dunn 