Entropy and its Implications for the Discipline of History
by
Rodrigo Fernos
Introduction: History as progress
Historians tend to explicitly reject the 'scientific model' for history while at the same time imbuing their characterizations of the discipline with a sense of progress. History per se did not truly exist until the 19th century, in which a more rigorous approach emphasizing primary sources was first called for and adopted (i.e. 'as it is', von Ranke). While various foundations were laid numerous times before then, dating all the way back as Thucydides in the Greek period or innovations introduced during the Enlightenment period (such as Voltaire's rigorous History of Louis the XIV), history per se had not really 'begun' as a modern discipline. Professionally there were also no 'historians' per se, except those which had been commissioned to write 'official histories', obviously of a biased nature. The new rigorous academic discipline had its early beginnings partly with Ranke, gradually adopting a wider range of topics and approaches, epitomized with the school of the Annales. Certainly, there have been some side turns, as with the recent postmodernist historiographical movement which, in trying to free the historical voice, went to far. Nonetheless, the 'standard model', if one could call it such, is that of a general sense of genuine progress. Scientific advancements in fields such as archeology throughout the 20th century are also characterized as further contributions from ancillary disciplines. We may, in fact, readily acknowledge the many improvement which have been made to the methodology of history in the Annales school, as well as the its democratizing element in the inclusion of a broader range of topics from that which the traditional history, which essentially was a type of chronological political science, tended to be practiced. Some of these might include: gender, science, economics, and so forth.
However, this very-broad characterization of history's own history tends to be framed within a rather narrow time conception, typically within basic units of centuries. The period of 100 years tends to have a magical attraction towards historians, who tend to base periods in these basic units, regardless of whether the internal contents of the history being described has its own particular periodic demarcation points. Seldom is it seriously considered what the implications would be of considering broader base units, such as millennia or 10,000 year periods. Certainly, from a practical point of view, this initially would at first appear to be a fruitless endeavor given that few historians live to the age of 100 years, and little do we know whether our own civilization, whose threats of nuclear war have repeatedly placed the outcome of mankind into question, would be existent during that time. But if Charles Darwin taught the sciences to expand its time horizon into the deep past, historians should likewise (as regularly done by astronomers) also seek to broaden their conception of time (future and past), and seriously consider its implications of these deliberations for their discipline. Such is what this paper attempts to do, particularly with regard to the consideration of how the reality of entropy suggests different historiographical presumptions from those which currently are held by the discipline. This brief piece is only meant to suggest a discussion rather than to provide a definitive answer to the question it proposes.
The Existence of Entropy in the Human Realm
Entropy is a term first conceptualized during the later 19th century to described the observed patterns of physical phenomena, particularly heat exchanges or processes involving the transfer of energy. Lord Kelvin proposed the Second Law of Thermodynamics, which basically states that the universe tends toward greater disorder (entropy) at a macrocosmic scale. While limited scale processes, such as those found on the Earth, might appear to disprove the idea, one might validly point out that this valid observation does not take into consideration the eternal energy input into the system generated by the sun. Evolution appears to be 'evolutive' (progressive), in that animals tend to develop into more advanced and complex forms, merely because of this added energy input, which if diminished would radically alter the phenomena at hand. Simply put, to increase the amount of organization in a given system there needs to be an external energy input into it (i.e. the Sun). There is nothing such as a free lunch; a household that does not regularly receive a given amount of energy (cleaning) will rapidly decay into a disorganized heap of clothing and rotting matter.
The same principle applies to the ability of a society to store records, which are typically kept on cellulose paper which typically weighs a great deal, and which typically remains unread for most of its lifetime--creating relatively high costs for their preservation over the long run. In the immediate time period in which records are created, they tend to be of immediate and significant value hence stimulating their initial preservation. Contracts are safely stored to insure that its signatories abide by its stipulations. Bills and receipts are jealously guarded should a consumer need to return a product that did not function as advertised. Laws, deliberations, and hearings are archived so that citizens and lawmakers might gain a better appreciation of the spirit in which they were formulated and the problems they intended to address. For these and many other reasons, records have an intrinsic immediate value, whose significance gradually deteriorates with the passing of time. Hearings for laws which no longer exist lose their importance, contracts between parties which have long ago died no longer have relevance, and receipts for items which have either been lost or stolen no longer have immediacy. As time advances, the law of entropy gradually enters into the domain of public records. Households throw away receipts, law partnerships demolish old contracts, and states (as in Puerto Rico) blithely get rid of unwanted paperwork that might be harmful to immediate political concerns. The implicit social order which was stipulated into the pieces of written record is gradually eliminated. Everyone wins--except the historian.
From a 'macro-economic' point of view, it also does not make much sense for a state to invest its resources into the long-term preservation of records, particularly so for the small state which tends to be characterized by limited resources (Puerto Rico). Dusty archives tend to remain dusty archives, unless the historical debate should suddenly find a need for those previously considered extraneous or unnecessary--which is seldom the case. The state tends not to be ruled by historians but rather by crafty politicians whose principal aim is to remain in power by addressing the needs of popular or elitist constituents which will back them up either with broad mass appeal or deep financial pockets. It is simply impossible for the state to maintain all of its records, and it will generally have to select those which it deems important, which inevitably (again) introduces the concept of entropy into the information sources of the state. There is a loss of information in the system (archives), which, by definition, implies a loss of order in the system (society). Banking checks which were made to certain accounts no longer appear, and hence difficult to verify the legal transfer of funds between parties. Local banks, such as Banco Popular of Puerto Rico, will only store its customers records for a period of five years. Given its millions of customers, and despite its computerized resources, it does not remain cost effective to maintain those records. Again, it is seldom the case that financial or political institutions be ruled by the dictates of historians, and it is seldom the case that the general population understands the particular needs of historians or the particular importance of given events. There is no immediate economic benefit to the writing of history, and hence other values tend to dictate the existence of the resources on which it is based. In other words, societies (general public) tend to be chronologically flat.
The Myth of Objectivity as a Function of Time
What does the existence of entropy, therefore, imply for the historian who so deeply relies on such public records for the writing of histories? It suggests that, to some degree, Thucydides was right. We may imagine the historical dynamic—the relationship between event, document, and historian—in the following manner.
The typical characterization is that historical objectivity (rigor) will increase soon after the event, and with due reason. The principal basis for this claim pertains to the emotive environment of the events; the recency of such events suggest that it is typically difficult to observe or analyze all features of a given event because these are implicitly deemed 'unapproachable' topics under the given social atmosphere, or simply because actors were too emotively affected by the incidents for a genuinely reflective historical analysis. In this sense, the quality of the study will certainly increase soon after the event, demonstrating the increasing side of the typical Gaussian curve. These experiences have provided the basis for the typical presumption which currently abides in the historical discipline.
However, if one considers the actual generation of records, and the quality of the history that is thereby produced, what emerges is a typical bell curve, where X is the axis of time (T) and Y is the axis of # records and/or quality of the historical treatise.
BELL CURVE
(NOTE: NUMBERS NOT EXACT DISTRIBUTION)
When an event occurs, say at point x,y (0,0), there will be a low point of information, generally speaking. When, for example, the atomic bomb in Hiroshima exploded, there was a small amount of documentation which existed at the time of event relative to the total volume which is eventually produced. This initial would typically have been government records planning the event, of a confidential nature. However, after the event, the amount of material drastically increased, x,y (1,1) (2,2), (3,3). Government studies were made, memoirs were written, international attacks issued, and scientific analysis of the post bomb scenario were done. Some amount of material will be classified, and will remain classified long thereafter, suggesting a continually growing curve, x,y (5,5). However the information reached a certain point, say 50 years later, when this information began to decay gradually, x,y (4,6). Naturally, the persons involved in the event--Truman, Oppenheimer, Bethe, Teller, and others--pass on (die), as is in the nature of things; which, for the historian, implies an irreversible loss of valuable information that can never be recovered, x,y (3,7) (2,8). The US currently has the capacity to retain much of its documentary information, and hence much of the original information still remains available. However, general features of the curve can be observed over a relatively short time span of approximately 50 years.
Yet if we expand our perspective into a broader timeframe, we notice that the tendency will be for an inevitable and substantive loss of historical information—dynamics which do not typically enter into serious consideration by the North American historical professional. For our particular perspective, it might be noted that the environment under which the history profession in the United States has operated has been rather anomalous. The surprising wealth and economic status in the immediate post WWII era made the US the global superpower, which although was challenged by the USSR during the Cold War period, succeeded as global 'hegemony' into the 21st century. This tremendous amount of wealth permitted that large sums of resources, small relative to the total percentage but large as a result of its absolute size, went into the social sciences and the humanities. History obviously benefitted from this wealth, but which distorted its operating presumption of the resource allocation of history in a given society. Although all societies keep some records for posterity, the vast majority of societies do not keep all records indefinitely. Record allocation and loss are inevitable processes which US historians have tended to avoid given their anomalous favorable social circumstances of the post WWII period.
It is far easier to observe the bell curve of historical records in small states, with less resources available, and hence the greater probability of resource loss (entropy). Puerto Rico might be a typical case. While doing historical research in government archives pertaining to one of the most important decisions in the last 50 years, the purchase of the Puerto Rico Telephone Company (PRTC) in 1974 by Rafael Hernandez Colon (PPD governor) for which some $350M were allocated and a projection of a $650M investment was expected, few records were found. Two sets of important and numerous committee hearings, one by Hernan Padilla in 1973 and another by Luis Izquierdo Mora in 1974, have vanished in thin air. There are more known historical records than actually exist. While certainly this loss can be attributed to the 1998 sale by the PNP administration of Pedro Rossello, and the likelihood that political pressure entered into the facile elimination of documentation which would have supported an oppositional party and/or stance (and lack of the rule of law in the island), the incident also reveals dynamics which tends to occur (entropy) with regards to historical records over the very long run.
Dynamics of an ideal world
Imagine, hypothetically, a self-sustaining society which by the very nature of its ideal sustainability is able to maintain a delicate balance to the environment, allowing it to exist 'indefinitely'. Let us also assume that, by definition, this society would have maintained a stable population base, a presumption that is a by-product of its self-sustaining character. What would the dynamic be under this hypothetical scenario pertaining to the relations between record storage and the social resources allocated to sustain these?
At the beginning of the history of this idealized state, the number of records would be very small, characterized perhaps by a small sphere. But, given that time goes on indefinitely, and given our condition of an idealized self-sustaining society, the material would also continue to grow indefinitely, eventually out of proportion to the resources available to that society. As in the relationship between surface area and volume, one growing at square rate rather than cube rate (x2 versus x3), the total volume of historical resources (primary resrouces) will tend to grow far faster than the existing population and/or economic resource base with each generation creating its own history. The sphere of information would reach a maximum 'carrying capacity' visa-vie the resources of hypothesized society; the total amount of historical information which that society can hold would have reached its maximum size.
TOTAL PRODUCTION OF HISTORICAL RESOURCES
However, it gets worse. Given that the information produced (historical records) would continue expanding after a certain 'maxima' point (given our condition of self-sustainability), the society would be forced into a position of having to throw away large segments of the available information. The more successful a society would be in maintaining a state of self-sustainability, the greater the predisposition would be toward the elimination of this information which, while not necessarily suggesting a loss in the absolute quantity of information available , would rather imply that the quality of information would begin to degrade as important segments of this body would be necessarily discarded—the downside of the bell curve. This would be the inevitable result of the need by that society to be forced to eliminate information (which is eternally being generated) visa vie a limited set of resources in which to contain this eternally-increasing information set.
EXCESS GENERATION OF HISTORICAL RESOURCES
Moral factors as insufficient criteria: Puerto Rico
An example that might help to demonstrate these dynamics, however, might be made from societies that have risen and fallen over a long time span, particularly the case of the Aztecs or the Incas. As numerous authors have shown, these expanded to the maximum point which their technology could sustain. The geographical expansion of their empire, akin to the increasing quantity of historical resources in our hypothetical case study, reached a maximum point, which thereafter began to decay to a point of social collapse. The arrival of Pizarro and Cortes merely sparked the beginning of processes that were ‘waiting to happen’.
While the case of Puerto Rico is afflicted by historical anomalies (political parties as competing economic sectors, hence lack of a rule of law), it typifies in real terms entropy dynamics of the most ideal society. Under the most ideal of conditions, even the most ideal of societies would not be able to retain all of its records indefinitely, but will in fact reach a point where historical primary resources will begin to decay. The process of historical entropy would eventually reach a point, where historical material would have to inevitably be discarded. The reason is that, despite the fact that the population maintained a self-sustaining relationship to the available energy resources and/or environment, (say a freezing of the y-axis in terms of population growth) the time x-axis would continue growing indefinitely, creating a mass of material disproportionate to its limited geographical size and economic base (even while presuming that these remain stable).
If the ideas here presented are valid, it suggests that the great loss of archival material in Puerto Rico that has occurred in recent decades is not merely the result of cultural particularities, as tends to be used and explained such as irresponsibility, ignorance, etc. (Moral explanations tend to erroneously be the first explicative step for phenomena. However, one explanation does not preclude the other.) Social sectors in the island are in a continual battle over resources, making the island an ideal laboratory of the processes invovled. (The limited rescourses visa vie the population base would put the pressure of natural selection towards the elimination of documentation, or towards an increase of entropy in the system.)
It is widely commented in private that many resources have been thrown away, such as a when governmental institutions are restructured, which has often occurrred in PR during the latter half of the century (ie Departamento de Obras Publicas, Comision de Servicio Publico, Autoridad de Fuentes Fluviales, Puerto Rico Telephone Company, etc.) Many important architectural designs have been lost, for example, when plans were transferred from one locale to another. The architectural plans of the School of Tropical Medicine are nowhere to be found. During an early phase of investigation, the author went to the ‘library’ of the Colegio de Ingenieros y Agrimensores, only to find that the very rigorously and elaborate card index, registering all of its magazine articles, were being used as scratch paper for a library that had been turned into a child-care facility. Countless other such examples could be mentioned.
These and many other cases suggest that the island is symptomatic for dynamics at a broader scale, as in the processes of island biogeography where rates of change occur at a faster rate and hence more readily observable.
Contemporary history as the only valid history
It is not being suggested that historical knowledge will ultimately collapse. The important element to point out is that the typified conception of the totality of social production of history seems to contain a number false premises that, in the long run are self-injurious to the profession. The presumptions of the ‘standard model’ of history, which fail to take into consideration the dyanamics previously described, might actually move the discipline over the long run into a state of decay over a very long time span. In other words, the discipline of history may not progress after all.
One such conception which could be questioned is the casual regard to which secondary works are typically treated. It is presumed (under certain circumstances validly) that the natural cycle of academic production will render secondary works obsolete, and hence a reduction in their value whereupon librarians are given the professional power to eliminate this should the need exist. It will be generally recognizedthat all fields of history have made tremendous improvements, generally speaking, over the course of the twentieth century, an example being older ‘positivist’ ‘documentarist’ histories.
However, the idea of an ever improving historical methodology, while valid for much of contemporary historical production, is invalid over the long run because of the false presumption of an eternally lasting absolute set of primary resources (as in the natural world). While today we may certainly be surprised at the unsophisticated methodology of the positivist historian during the early part of the century whose historical tracts tended to be compendiums of documents and other primary sources, the guiding asumption of this critique is the current availability of the resources on which earlier historians wrote their studies. The disappearance of the underlying primary sources would suddenly give them a great deal of value, in contrast to the despective amusement in which they are currently held.
At certain periods of history and in certain societies, certain historians will have access to the maximum range of primary sources available in (literally) an eternal historical timeframe, and will produce 'secondary works' of genuine historical merit—much more so than the primary sources on which they are based given the synthetic character that all histories contain. They will genuinely represent a 'universal literature' whereupon attempts to study the same topic by historians in later later will be doomed to be of lower quality given the inability of these later historians to access the full range of resources which former historians had available. This suggests that societies by nature have an need to create a 'canon' of some sort that validly targets these maxima points on the historical continuum of certain specified themes or topics. The canon will be the conservation of non-repeatable, irreversible, analysis in the sense that these can never be duplicated; itself caused by the inevitable decay of primary sources.
In a simplified manner, it is being suggested that the Thucydiean approach to history is perhaps the most appropriate one in the very long run. That is to say that "contemporary" history ultimately be the most valid history of all. To repeat, at some point after the event "contemporary" historians will reach a certain maxima of resource availability which will allow them to produce the fullest and most genuinely comprehensive histories of their respective subjects, which will not be reproducible in later periods. The preservation of these 'maxima' will be the most rational long term policy of all, if the aforementioned conditions are accepted.
It should be noted that a canon, as conceived under this structure and definitions, will however not refer to one or two idealized individual historians writing about a given particular topic (as has typically been done), but rather will likely tend to cluster around groups of historians in given periods. Records, after all, do not just suddenly disapear but rather tend to exist for quite some time, depending as previously noted on the immediate needs of that society for such records, which might legitimately even include the writing of certain histories, but not history in the most comprehensive meaning of the word. Their existence is diffused and spread out over various chronological set points.
The principal problem that arises is that of determining the “points” of maxima and minima along the historical continuum, which will likely be subject to attempts by ideologues to tyrannize the historical profession to propose a certain views of history. Aside from the problem of subjectivity which naturally arises in the typified case, a more serious problem arises out of the very nature of the time scale that is here being presented, and out of the uncertainty which all individuals have pertaining to the character of the future. The same problem exists with regard to predictions of the duration of non-renewable resources as petroleum (i.e. Huppert's Peak). While these also can be typified by a bell curve, it is difficult to trace the scale and size of the curve, and (more significantly) the point at which one currently resides. Despite recognition of its validity and presumption, it is hard to put into practice, to actually determine the maxima points on the historical bell curve, given the numerous uncertainties involved.
In the real world, outside the boundaries of a hypothesized idealized society and condition, none of us know how long our resources will last, how a nation's economic system will thrive, or whether a foreign enemy at some point will occupy our borders. This is impossible to foretell, just as it is impossible to predict where all atoms will be situated at any point in time with the most sophisticated of computers. For the purposes of this essay, it might be specifically asked how far away from the date of an event should the points of maxima be established? Should the time scale be of 25 years, 180 years, 3,000 years? What features of the historical process could be used as markers to help identify these points? How would they be related?
The theme raises more questions than it answers. These are questions difficult to answer, and which make this brief essay also only a hypothesized cognitive inquiry rather than than attempt to present solutions to its proposed problems. Perhaps algorithms could be established which would help determine these natural flows, but such would be beyond the scope of this brief essay and the ability of its author.