Retread Resources Blog

Epistemology— Truth as Related to Science:

I will again start with a definition of some terms related to this topic.
From Wikipedia:

“Epistemology (from Greek ἐπιστήμη – episteme-, “knowledge, science” + λόγος, “logos”) or theory of knowledge is the branch of philosophy concerned with the nature and scope (limitations) of knowledge. It addresses the questions:
•    What is knowledge?
•    How is knowledge acquired?
•    What do people know?
•    How do we know what we know?

Much of the debate in this field has focused on analyzing the nature of knowledge and how it relates to similar notions such as truth, belief, and justification. It also deals with the means of production of knowledge, as well as skepticism about different knowledge claims.  The term was introduced into English by the Scottish philosopher James Frederick Ferrier (1808–1864)

Truth is a commodity and can have a variety of meanings, from the state of being the case, being in accord with a particular fact or reality, being in accord with the body of real things, events, actuality, or fidelity to an original or to a standard. In archaic usage it could be fidelity, constancy or sincerity in action, character, and utterance. The term has no single definition about which a majority of professional philosophers and scholars agree, and various theories and views of truth continue to be debated. There are differing claims on such questions as what constitutes truth; what things are truth-bearers capable of being true or false; how to define and identify truth; the roles that revealed and acquired knowledge play; and whether truth is subjective, relative, objective, or absolute.

Belief is the psychological state in which an individual holds a proposition or premise to be true The terms belief and knowledge are used differently in philosophy.

Epistemology is the philosophical study of knowledge and belief. The primary problem in epistemology is to understand exactly what is needed in order for us to have knowledge. In a notion derived from Plato’s dialogue Theaetetus, philosophy has traditionally defined knowledge as justified true belief. The relationship between belief and knowledge is that a belief is knowledge if the belief is true, and if the believer has a justification (reasonable and necessarily plausible assertions/evidence/guidance) for believing it is true.

A false belief is not considered to be knowledge, even if it is sincere. A sincere believer in the flat earth theory does not know that the Earth is flat. Similarly, a truth that nobody believes is not knowledge, because in order to be knowledge, there must be some person who knows it.

Later epidemiologists, for instance Gettier (1963) and Goldman (1967), have questioned the “justified true belief” definition, and some philosophers[who?] have questioned whether “belief” is a useful notion at all.”

In my essay Truth, Consequences and Reality, I have already referred to Plato’s Theaetetus in parable form.   I am not qualified and have no great desire to conduct, a long and detailed philosophical underpinnings of this or Epistemology in general.  I don’t think you, the reader, desires to read one here.  In other essays of this series, I have assumed the reader is acquainted with the literature, here too this assumption will apply.

Anyone doing or thinking about science brings to that action or thought a preconceived Epistemological model.   Each of us, depending on our education and socialization, will have a model that is different in detail but I contend, in general much the same.   Those of us who are purveyors of science, fancy ourselves as seekers of truth.   Like Socrates we have “this much knowledge” and nothing more.  Unlike him we are not comfortable in our state common ignorance.

Diminishing that state of ignorance is a major driving force behind our scientific efforts.  Having reached some state of enhanced knowledge, i.e. added some “truth” as we understand it, we publish our results.  Remember, in using the Scientific Method, we do not prove we can only disprove or through experiment verify.  Remember also, the Scientific Method is itself a model.  The application of that model is influenced by the Epistemological model we scientists bring to it.  In the most simple case, we have a model that influences another model.  In the real world, I would suggest it is many models, all influencing each other, a matrix of models.

While our Epistemological models may not all be exactly the same, they are sufficiently similar that agreement on what is truth, as applied to objectively collected data, should be attainable.   Basic Meta data requires a more rigorous examination.   To keep this discussion simpler I am dividing Basic Meta Data into two divisions.  Division One is that meta data developed on a data set derived from a single location or multiple equatable locations, to which most simple and standard formulas or operations have been applied, i.e. mean, mode, standard deviation and so on.  Truth here is the voracity of the raw and first order meta data. (see the essay on data)  Agreement here should be achievable.

Second, third or fourth order meta data become more problematic.  My Epistemological modes requires that these data be demonstrated to produce the precision and accuracy claimed.  Notice the adverb claimed.  If the presented meta data is not fully documented and checked, shown to meet the precision and accuracy supposedly achieved, by the more sophisticated operations used, they must be rejected as unreliable or inappropriate.   Like it or not, the data and its voracity are the very foundation of science.  It is not the hypothesis or theory or idea and never the model being applied but the data set that is of ultimate importance.   The only true knowledge and the only sure truth, what ever that is to be, lay with the basic, i.e. underlying data.  The base or foundational data set(s) must always be preserved in a raw and completely untouched fashion.

Raw data sets often need more that a little adjusting to be useful in any kind of analysis.  Starting from the premises, all I know for sure, i.e. all I know to be true, is that which constitutes my raw data set, we find ourselves greatly constrained.  My Epistemological model allow for first order meta data.  I can then produce from the base data a second basic data set as defined above.  My model allows that to be called truth.  Having properly documented and accurately preformed those adjustments, which now can be verified by others, they too can accept them as truth.  Example:  hourly temperature readings at a weather station(s) from the base data set.  TMax, TMin, T Mean and so on the first order meta data for each day and so on for each data location.  As long as the methods used are agreed upon (formulas, conversion factors, etc.) and the arithmetic correct, all should be able to agree.  This data set too, must be preserved.  What that data represents, in addition to itself, must ultimately be determined.  Great care and thought must be given to how data from disparate sources, different types of measurements and so on are compared and coupled.  Historical or “fossil” proxy temperature measurements are fraught with difficulty, for example.

It can be problematic when further adjustments, refinements or constraints to the data set(s) are being applied.   The application of any adjustment factors applied, to any or all this truth, produces those second, third and forth order meta data sets.  These data sets must also be preserved but now have suspect voracity.  They may be called data in the vernacular, something I myself often do.  They are not true data and can no-longer be classified as truth.  Almost any further adjustments, beyond the first order meta data, results from the application of one or more models.  Models are not reality and models can not produce data, only results of application.  Therefore, I must restrain myself and call this new information results, i.e. meta results.

My Epistemological model can not accept meta results as truth.  It can accept them as model results, interpretations and so on, but never as truth.  Others may have different views, the sophists and propagandists for example.  That is acceptable as long as they clearly identify exactly how and what was done to produce them.  I would suggest, why it was seen as necessary to make the adjustments, should also be clearly defined.  This may be considered a lot of extra effort by the researchers but it is necessary.  Those of us in the business of doing science must attempt to be objective in what we do.  Errors will occur.  When found those errors must be corrected.  At no time can data or meta data be adjusted to force a fit against or toward, the expected results of a model, theory or idology.   If corrections or adjustments are required by the modelers, those actions must be completely documented and the results of those, now significantly hight order meta data, identified as such.  I would also suggest that the difference between the two results sets also be clearly shown.   The present controversy relating to how climatological average temperatures collected, since people started recording these things, is a perfect example. (see the discussion on precision and accuracy.)

Each and every step, in each and every part of the scientific experiment, must be clearly identified and duplicatable by others.  Mathematical models, such as those used by climatologists, are nothing more then un-calibrated experiments.  In the case of any models or sub-models being applied to meta data sets, I suggest the computer code also be readily available to any and all examiners.  This becomes critical if different methods, constants or factors are utilized. (see the other essays in this series.)

Truth then is a bit like beauty, in the eye of the beholder.  It does have some objective qualities.  Results, i.e. higher order meta data can be true, in being accurately produced by a set of rules.  The problem here lay not in the definition or the voracity of the calculations.  It lay in the objectives and reasoning behind those adjustments.  While we can all or mostly all agree that a measurement or set of measurements are true, we are often, not in full agreement with adjustments subsequently made, to that set of measurements.

If I happen to be the one making those adjustments, for what I understand to be good reason, then my Epistemological model tends to readily accept them.   If you follow a different logic and apply it to the same base data, I may or may not accept them.   My model is highly suspicious to broad based adjustments but less so for data point by data point corrections, on a case by case basis.  Historical temperature data adjustments is one example.  Shifting UTM locations between Nad 27 and Nad 83, is quite another.  In the case of the UTM’s, each data location is adjusted from one model to another model.  In this example the data is still in the base or True mode.  It is locational information and can be verified in the field, using either coordinate system.  If a mark was put on the ground using Nad 27, recalculated and revisited using Nad 83, the mark would not have moved in a physical sense.  No interruptive influence will be seen.

If a broad based adjustment is made to a large subset of the temperature data, some interruptive change should be expected.  In the former case the geographic datum is changed, in the latter the interpretative datum.  In both examples the changes were broad based and relative to mathematical model.  In the latter the influence of the change in subsequent modeling efforts could easily produce interruptive ramifications.  As we see, the raw data in the locational example is not changed, it is only adjusted to a new datum.  The XYZ location of every point is mathematically different.  The change has been universal and relative.  In the temperature example, the base or raw data was not universally adjusted but only a sub set of it.  This is no-longer true data or even first order meta data.  It is now a mix of true and higher order meta data.  The voracity of the entire data set has been compromised.  It can now only be a result.

In the recent past and the present, several large data bases have been and are being used by scientists around the world.  In one instant the authors and data base keepers have refused to make the base, unadjusted data available to other researchers.  In the other, apparently the keepers have made high order meta data “adjustments” without informing the users or explaining just what was done and why.  I am not sure if and for how much longer the base data or the first and second order meta data will remain available.  In the former case the data base keepers were obviously motivated by political and personal reasons.   This is not how science is done and it is not how science is supposed to be done.  It also appears that numerous tweaks were made to the higher order meta data that were not or poorly documented.  Some are calling this “Climategate” in the press and bolgisphere.

As to the fiddling with or homogenization of temperature data, or the changing or fiddling with what was supposed to be raw data, the situation is so new I can not comment.  I will say that this may be real or may be a tempest in a teapot.  I suspect part of each.  I have seen examples of this, so something is amiss.  That going on may be quiet forthright or may not be.  I strongly suspect the simple rules I suggest above have not been closely followed.  This may all be more communication gap then nefarious.   The problem here is that many researchers and huge expenditures have been applied to using these data sets.   If the voracity of that data is in question so is any work that resulted from using it. (see essay on Trust.)

What motivation would a rational practitioner of science have, to in any way violate the very spirit of the scientific method?  My personal view is ego or economic self interest.   Admittedly that is both cynical and inadequate.   I don’t personally know any of these people on either side of this or any other question.   The ego driven and economic self interest is self evident.   I do know, that for a very long time now, a difference has existed between science and faith or ideology.   The history of science is filled with examples.  The major difference between the two world views is simply: when new evidence comes to light that casts doubt on orthodoxy or present hypothesis, the hypothesis is dismissed or revised.

In ideology/faith, if such evidence is presented, it is censored.  Galileo presented the Papacy of his day with just such a challenge.  (I know it was much more political and complicated then this but that is all beyond the scope of this essay.)  The idea that the earth revolved around the sun was heresy to the Church, since it took man away for the center of the universe.  That position was ancient Greek and therefore pagan.  The Church had invested in this notion.  It was dogma and the orthodox view.  That investment must be protected.  It smacked the collective ego of authority.  Similarly to draw attention to the inadequacies of the present climate models and the inadequacy of the modelers understanding of the processes, therefore their conclusion and predictions, removes man as the center of the climate system or climate universe.  Suggesting that perhaps man has fewer sins to atone for, must therefore be stifled.

I have spent a goodly number of hours reviewing all the books and writings on philosophy that exist in my library.   I could not find any reasonable justification for the end, however nobel, justifying the means.  I was reminded, this is a world view well articulated by Machiavelli in The Prince, although he never quite states it as such and it is a much older concept.  His work is less philosophical then political.  To paraphrase: if the Prince is to maintain his position or strengthen it then… what ever it takes to do so is justified, since the Prince’s desired outcome is defined as the greater good.   While he may have addressed his work to a Prince of Florence, he could just as easily addressed it to the Prince of the Roman Church or any other leader of the day.

This kind of thinking has no place in science, regardless of the name give to the study of the political.  No model can produce results so important to humanity that unscientific or immoral behavior can ever be justified.  In copenhagen, in the recent past, we saw a most tragic misuse and misunderstanding of science.  The sophists, propagandists, politicians and advocates appeared to masturbating with the orthodoxy of climate change i.e. AGW.

The half truths, outright lies, misunderstanding and hubris overwhelmed this writer’s sensibilities.  What I saw was a desperate attempt to bend the Faith of anthropomorphic generated climate change to political and economic advantage.  Mostly it failed.  This is not a defeat or victory.  It is just another way station along the arrow of time.

My understanding of earth science, my model, suggests that change is always with us.  It has always been with us and will always be with us.  Change is normal, the static is not. This is not my eureka moment, its roots are as old as the western philosophical traditions. I do not know for sure if man’s activities impacts on natural change in more then a minor way.  I do know that if the earth was not dynamic and always changing it would look much more like the moon than it does.

I do know that average the temperature of the atmosphere has been rising since the coldest times of what we call the Little Ice Age.  I know that it is simple and tempting to apply some kind of linear rate to any data set.  I know this kind of thing has happened before before modernity was even conceived or humanity for that matter.

I have often said among my friends, we humans choose the fiction we wish to believe.   I have no problem with that.  I do have a problem when anyone foists that belief on me or others, as anything other then just that, his or her faith.  The world views and objectives of faith and science are different, in that they ask and answer different questions.  Faith address why and science how, mythos and logos.  Society is best served when these two questions are kept separated in our politics and economics.  We have a hope of establishing the truth about how, we have no hope of establishing the truth about why.