Monday, June 8, 2015

Arguing in Hindsight II

NOTE: This article has little to do with the original "Arguing In Hindsight" other than highlighting a very similar error in reasoning. 

There was, at one time, a popular theory that governments, or perhaps cultures or empires (take your pick), were, like people, subject to a natural lifecycle. That is, empires would be born, would mature, would go through an explosive, expansionary youth, settle into a complacent middle age, and then gradually decline and die.  It has been expressed in a number of different ways, justified by a hundred different arguments, but basically it is little more than a direct analogy between states of some sort and organic life. In fact, it is not even a wholly extinct myth, though not as common in academic as it once was, it still pops up from time to time in informal discussion, and, at times, even gets smuggled into popular histories and political science.

The problem with this theory is quite simple. Though supported by a few complimentary errors1, at its base, this theory only works because we are arguing in hindsight, looking back at the events, and then fitting them to a pattern. I shall argue that the theory itself is in error, based as it is upon cherry picked examples and forced analogies, but, even were the theory correct, there is a second problem, that it is of no use in interpreting current events. As I said, it is only possible to fit history into this mold in hindsight, we can only tell when a nation hits old age because we know when it died, but, looking at the nations of the present, it is impossible to tell which are old, which young, and so on. But we will discuss that shortly, for now, let us look at this theory and see how it demonstrates the errors inherent in arguing in hindsight.

To make my case, let us start with a method familiar to junior high Geometry students2 everywhere, the indirect proof. For those who don't recall their math classes, or only remember drawing lots of arcs and lines using a compass and straight edge, the indirect proof was The Other Sort of Proof, that is the one that didn't involve a combination of line drawings and double entry accounting (at least that is how direct proofs always look to me). The indirect proof was the clever one, the one where, to prove A was true, you started by assuming A was false and then, through a series of logical steps, showed why that assumption produced some sort of logical impossibility, usually by finding some contradictory conclusion. And in this way, you would prove, at least to the satisfaction of geometry teachers, that A must be true3.

So let us start by assuming there is some sort of defined life cycle for governments, or at least for states (some say for "empires", but since my examples will be both states and empires, that is an irrelevant distinction). Even if we cannot identify the point at which we are living from the evidence around us, presumably, in hindsight, it would be possible to tell what stage a state has reached, say when it leaves adolescence, or reaches middle age, simply by knowing the course of its entire history. And so, if we were to take a number of states, empires, government, what have you, and lay out their historical path, we could see clearly that they progress from infancy, through childhood, to adolescence, early adulthood, maturity, middle age stagnation, and on to eventual senescence, decline and death. If the theory holds true, and if we exclude those cases where outside factors prevent us from following the entire course -- say those states destroyed in their youth, or other cases where outside influences prevented the natural course of history from being followed4 -- then all the remaining states should present a clear picture of normal growth, the way a portrayal of the lifespan of a number of human beings would show a similar progression through the normal stages of growth.

However, if this were not the case, if we could find a number of cases where, despite a lack of outside influences, despite being allowed to grow normally, states were shown to follow a path unrelated to the supposed normal progression, then it could be assumed that we had shown the theory was false. Granted, one or even two examples may not prove the case conclusively, since it could be argued there was some influence which altered the normal course of affairs, or some factors we could not see which twisted the normal lifespan of the state, but if we can provide a number of examples, well eventually the claim that there was some unseen impediment would begin to sound hollow, and it would be pretty clear the theory simply does not match the reality.

But there is a problem even here, as even deciding where to being selecting examples and counter-examples raises questions the theory does not answer easily. For example, there is a problem even defining the basic terms, especially deciding what constitutes a state or government? For example, are the Anglo-Saxon kings of England part of the same state as the kings after the Norman Conquest? And, are those Anglo-Saxon kings of England part of the same state as the preceding kings of Wessex? A bit farther afield, how do we consider the kings of Mercia and Northumbria and East Anglia and Kent and so on? Are they part of the same state? Or rival states which were absorbed? What of states such as Hungary or Bohemia, where the royal families merged together two previously separate states? Is the new state separate and thus starting over in infancy? If not, where does it come in the development of the state? And what if the two merged states were of differing ages, on very young one old? How do we conceive of the merged state? What is its age? And what of states that were overthrown, but whose conquerors claimed to continue the tradition of those states? Is the Ostrogothic kingdom of Italy independent and new? Or a continuation of Rome? And if either, how do we think of the preceding semi-nomadic Ostrogothic state? And how does the Holy Roman Empire relate to this pattern? Or the government of the city of Rome itself (or of many other Italian cities) which survived the fall of the western Roman Empire and persisted, even though many states rose and fell around them? Or how about France? Was the revolutionary government a continuation of France or a new state? And what about when it became the Consulate? Or the Empire? And if those were new states, then what of the Bourbon restoration? Was that yet another state? Or a resurrection of the prior state?

I could go on, but the basic problem is easy enough to define. How do we define a state? Or, to elaborate a bit, when does a state start? When does it end? When does one state become another? What of subordinate governments that survive the state that used to control them? How do we understand them? And what of states that merge? States that split apart? In short, if we are to make an analogy from a living organism to a state, how can we do so when the state can do so many things completely alien to living beings5?

This difficulty in marking the start and end of states is a problem in more ways than one. For example, it undermines almost any argument, as, assuming you disagree with a conclusion, you can simply dismiss it by claiming it is based on an improper understanding of when the given state began and ended, or similar details. For example, two of my arguments to come are based upon the British Empire and the Byzantine Empire. However, it would be incredibly easy for someone to dismiss my claims by arguing I defined the starting and ending point of either empire improperly, or was viewing the various periods in the wrong light, calling adolescence adulthood or vice versa. The fact that both have a long and involved history, that they were involved with other states, had various government come and go, saw considerable changes in their rulers, their territory and their traditions, all make it possible to argue for many different points where they began, reached their peak or ended. And thus, in many ways, all of this argument is futile.

But, before I convince myself my entire essay is pointless, let us start with a relatively uncontroversial example, one which has pretty clear beginning and end, where there are no subsidiary internal government to get in the way. Where there is no successor state claiming to continue the legacy. Where the sequence of events, and the various peaks and troughs are pretty well known and largely beyond dispute. And finally where the state in question no longer exists in any similar form (or really any form at all6) making it clear the state has lived its entire lifespan, and thus making it clear we can discuss the entire cycle, we don't have to worry if the present is still adulthood, or might instead represent middle age or even dotage7.

My example is the city state of Athens during the classical period, and beyond, and it strikes me it is one of the few examples about which this claim can be made. But, since it is one of the few states providing such an ideal fit, let us use it to look at how well -- or poorly -- the thesis fits observed facts.

There is a small problem of defining the lifespan of the state, though not as much as with many others. Athens had a continuous existence as an independent state from its birth in the distant past, thorough to at least the reign of Alexander, or, if you want to view that charitably as an occupation, not a conquest, then all the way through to the eventual absorption into the Roman empire. However, as that definition suggests, there is a small issue. At least twice in its existence, Athens was occupied and had a government forced upon it. Though, in the first case, the Thirty Tyrants imposed by Sparta, the rulers, though pro-Sparta, were still locals, and ruled for a rather brief period. Alexander did a more thorough job of subjugating the city, but it too was short-lived and, by the early stages of the successor states that followed his death, Athens was effectively independent again. So, though I suppose one could argue for Athens ending with either occupation, it seems to me that it makes the most sense to examine the city's lifespan as a period running from its founding until the eventual rule of Rome. (Actually, as we shall see, if choose one of these imposed government as the death of the Athenian state, it actually makes the case worse for the "state as man" position.)

The problem with viewing Athens using the "state as a living being" thesis is that Athens had a tendency to swing from wild success to crushing failure quite rapidly. Rather than following the gradual progress the theory would promote, rising to prominence then slowly declining into eventual collapse, Athens went from the the early victories of the Persian War, to some rather substantial defeats, followed by some dramatic triumphs. Similarly, we see Athens rise rapidly to an empire leading the Delian League,  being followed quickly by tremendous Spartan victories, reversed with equally massive Athenian triumphs, only to have it end with Athens occupied and defeated. But again, within years, Athens rises once again, and, though not the imperial power it once was, it still manages to reestablish itself as a regional force once more, only to be occupied once again, and then, upon the death of Alexander, again repeated the miraculous revival, establishing itself as a player in regional politics, only to see final loss to Rome. In short, Athens followed nothing even closely resembling the supposed accelerating rise followed by decline and loss the theory would suggest.

Nor is Athens in any way unique. Most powers have followed an equally chaotic path. Byzantium enjoyed unparalleled power in the eastern Mediterranean after the chaotic period following the collapse of the western Empire settled down, but then was almost destroyed during the early expansion of Islam. Fighting its way back, Byzantium again enjoyed considerable power, only to see it largely crushed once more, and to then have the crusaders of the fourth Crusade actually besiege and occupy Constantinople. But, rather then die, as the theory would seem to predict, Byzantium enjoyed a revival, admittedly as a much weaker state, retaking the capital and much of the lost territory, allowing them to survive the eventually invasion of Timur, and even to hold out for a time against the Turks. Again, rather than a rise and fall, the pattern seems to be, for Byzantium a series of rises and falls, without much rhyme or reason.

Obviously the counter-argument to this is, that by focusing on short term gains and losses, I am not seeing the larger picture, that my descriptions emphasize too much the temporary ups and downs, and misses the full pattern.

But that is why I have named this essay "Arguing in Hindsight". The pattern that is found in this theory, the supposed rise and fall, growth and decline, youth-adulthood-senility and death pattern, only exists in hindsight. It is only when we know which ups and downs to value and which to ignore that we can say "ah, yes, here they are in decline". At the time of the events, it is not possible to tell if a defeat is a sign of impending doom or a simple bump in the road. In fact, it is not possible to tell even after the fact, at least not until the end is reached, at which time we can then impose the pattern and make the facts fit. Otherwise, even after the fact, it is difficult to say whether a given moment shows a state growing from childhood into adulthood, or simply a lucky break, or whether a downturn is just a momentary setback or the beginning of the end. All is only seen by imposing the pattern after the end has been reached.

Even if the theory were valid, and I am not about to concede that, but if it were, it would still be of no use for making decisions in the present, or identifying current conditions. If a given event can only be identified in the distant future, then, at present, it is impossible to tell if a given state is rising, falling or simply coasting along, since a given event, for good or ill, may be meaningful, or may be irrelevant, and there is not way to tel which at the time. Thus, even if ti were to be shown to be true, it would be useful only in the distant future, when one can see the entire course and decide which events are significant. For vents in the present, or even the recent past, it is pointless.

Not that being pointless is sufficient to disprove the theory, there are theories which are true, yet provide us with no predictive abilities, but I would argue, in this case, the reason the theory is useless, that it cannot provide anything even approaching a prediction is precisely because the theory itself is invalid. It is only by retrospectively picking and choosing, by selecting which events to emphasize and which to ignore that history can be made to fit the pattern, and thus, the theory itself is nothing more than theorizing in hindsight, taking the past and fitting it to one's preconceptions.


1. This theory also has a few other errors, which we shall briefly discuss in the essay proper. For example, by cherry picking examples to fit the pattern, such as using Islam as an example of how "young" ideas explode into the political scene, while ignoring the first three centuries of Christianity, or the early years of Judaism, neither of which fit this pattern well without serious fiddling.

2. For some reason, I seem to be the only person who loves geometry. From 7th grade Alegebra I through Pre-Calc in 10th grade, everyone I knew -- at least those who did not despise all mathematics -- loved Algebra and hated Geometry. Even when I got to college, it seemed in "Calculus and Analytic Geometry" the emphasis was inevitably on the algebraic "calculus" parts, and very little attention paid to the geometric bits. On the other hand, I was not terribly enchanted with Algebra, but got consistent perfect scores in Geometry. (Yes, even on my midterms and finals. I was something of a math geek, at least when it came to Geometry.) And later in life, when I encountered Calculus, at first I had a terrible time with it. In part it was because of non-English speaking professors speaking to lecture halls without a microphone, and teaching assistants who knew only Korean, but there was something else. When I took Calculus I again, taking advantage of my the free tuition that came with my job -- determined to get a better grade in the one subject in which I had ever earned a C  (as I said, kind of a math geek) -- the fact that my teacher slipped in and out of Russian did not stop me from getting it, and the reason was that he used geometry to explain the concepts, rather than sticking to purely numeric algorithms. So, please forgive me if I sometimes seem overly fond of Geometry, but, for whatever reason, it is one of those subjects -- like assembly language, Economics and Latin -- which just clicked for me.

3. The indirect proof can be abused if you misuse it. I showed one way to do so in my proof that all numbers are equal (see "Shorthand and Confusion" for a discussion of that topic). Most often, an indirect proof can be abused by selecting a situation where A and not-A are, despite appearances, either not exhaustive, or not mutually exclusive. The first can be done by defining not-A in some idiosyncratic way, so that there are remaining items neither A nor not-A (such as assuming every non-Republican is a Democrat). The other is to pick areas where the division is not black and white,for example, say male and female. Traditionally these would seem mutually exclusive and exhaustive, but consider, say RuPaul or Bruce Jenner and suddenly we can see that, depending on how we define the terms, they may not be mutually exclusive, or, in the alternative may not be exhaustive. By using such improper pairs, the indirect proof will result in conclusions that may not be valid, so one should be careful.

4. Of course, I could argue that, as the normal lifespan of a state is inevitably experienced in a world crowded with other states, and the "natural habitat" of a state is a world where it competes with other states, to claim interference by another state is unnatural or prevents the natural evolution of a state does not sound right. I will grant, the destruction of one state by another may unnaturally terminate the progress of that state, but short of destruction, should we not consider interaction with other states perfectly natural and normal events, and assume the states involved would continue to follow the ordinary course of growth for a state?

5. This analogy, from the Greeks onward, has always been problematic, simply because a state is not a living being, and trying to make it fit always seems to make a mess of any argument. In short, it is simply a terrible analogy. But, fortunately, there are those who have made the young state/old state argument without explicit analogies to living beings, so the fact this is a terrible analogy is not fatal to the theory.

6. I grant the city of Athens still exists, and, with some brief exceptions, had a continuity from the time of Pericles (and before) to present (if we ignore the possibility of complete extermination during one of the gothic or slavic invasions, which might have been possible). However, since the city itself was subordinate to another power from the time of the Romans onward, it seems fair to take that conquest as a breaking point, where an independent city-state government ends, and is replaced by a civil government subordinate to an external power. And thus, I think I am safe making the case I did. (Though the fact even this case has some fuzzy boundaries to it shows just how badly defined the whole topic is.)

7. England is a perfect example of why surviving states are so problematic. If we consider the loss of empire as a sign of Britain's decline and senility, then we have to explain her continued existence and even periods of improvement. On the other hand, if we consider the present as late adulthood, or sometime in middle age, we have to explain the loss of empire without reference to decline, and explain how Britain could endure such a loss while still in her prime.

No comments:

Post a Comment