Search This Blog

Saturday, July 31, 2010

Is human finitude a reason to abandon methodological naturalism?

Before I get into the meat of the post, just a reminder to someone who happens not to have read the previous posts that I am continuing a series about methodological naturalism.  As mentioned in a previous post, I will not be using that term in favor of the terms "detectable, observable, and measurable" to describe the requirement in standard scientific method that theories be supported/confirmed by "detectable, observable, and measurable" data.  

Human finitude complicates the issue of detectability. If somehow we had made everything ourselves and had the design plans in hand, we could account for everything, even things we could not detect with our senses and instruments.  In fact, the world we know existed before us and will continue after we perish, we had no hand in devising the rules by which it operates and can control almost nothing that happens in it, and we find ourselves continually frustrated, surprised, and puzzled by the things we encounter in it.  In short, we are finite, derivative, and accidental.  Given our situation, many entities, events and processes are 
1.  in fact undetectable given our current limitations, or
2.  in principle undetectable given our derivativity, accidence, and/or finitude.  
Consider the following scenario:  The entire expanse and history of the cosmos we inhabit is encapsulated in the oscillations of a sub-atomic particle in a drop of coffee falling from the cup of an inhabitant of the outerverse sitting at her kitchen table getting ready for work.  The immensity of this outerverse is such that our cosmos's transition from big bang to thermal stasis happens while the drop is still falling.  Entities, events, and processes in this outerverse caused directly the initial conditions which led to the "big bang" starting off our cosmos.  Oscillations of nearby particles in the atom, molecular interactions in the drop of coffee, the gravitational forces pulling the particle and everything around it in the drop of coffee toward the surface of the table -- all contribute indirectly to the forces and entities we experience, and given enough time, some entity, event or process in the outerverse could lead to the sudden destruction of most or all of our cosmos. It must be kept in mind that the outerverse's scale of time and distance is so great relative to our own that even the approach and passing of another sub-atomic particle in the outerverse would take longer than the entire history of our cosmos. 

The immensity of this outerverse makes it undetectable.   There is simply not time enough in the entire history of our cosmos for us to accumulate enough evidence to be able to grasp the makeup of the outerverse or even detect its existence.  Furthermore, the laws of physics familiar to us do not hold in the outerverse's scale of time and distance, just as Newton's laws were found not to hold at the sub-atomic level. Finally, the inhabitants of the outerverse are just as unable to detect our existence or send us any kind of communication or message that we would be able to receive.

I bring up this scenario so that I can remove the issue of religion from the discussion of detectability and scientific method.  Believers in non-physical, non-natural, "spiritual" entities will object that this scenario leaves their perspective on the nature of the cosmos out of the loop.  To show what a nice guy I am, let's bring God into the picture big-time.  Since I am most familiar with and still favorable to Judeo-Christian theism, we will add the following:
1.  The outerverse is a direct, immediate creation of God.
2.  Although our cosmos developed entirely as a direct effect of natural causes, once human beings appeared God began to intervene via miracles (theists can add miraculous elements arbitrarily to this picture).
3.  Even though some kind of evolutionary process led to the appearance of life in this cosmos, including human beings, theists are free to add initial conditions that would preclude objectionable aspects of the evolution theorized to have occurred in our world -- parasitism, predation, sexual competition, etc.
4.  Humankind fell.  Theists are free to assign the same cognitive effects of this fall as they believe occurred in the "fall of Adam."
5.  God provided specific direct revelation about our cosmos via whatever mechanism theists want to propose.   He told humankind that He created the cosmos and everything in it.  But he specified no mechanisms, spoke nothing about the existence of the outerverse and left no instructions about what humankind could or should do to try to find out how He did it.

So here we have a thoroughly "supernatural" reality.  We  have one case, the outerverse, which meets only the minimum requirements of the correct defnition of supernatural:  entities, events, and processes in the outerverse are undetectable, unobservable, and unmeasurable.  In this case, immense time and space are sufficient to render the outerverse supernatural according to Arthur Strahler's definition.  On the other hand, we have a personal, "spiritual" God.  He fits the minimum requirements for being "supernatural" but in addition has the characteristics of intelligence and purpose.

In this kind of universe human science would always be incomplete in principle, not only in regard to an explanation of the human condition but even in regard to the origin, nature and history of the natural world.  Those outerverse entities, events, and processes would keep breaking the regularities proposed by scientists working purely with data discoverable in our cosmos.  Not that all scientists would accept this view of reality.  Some would strive for complete explanation of the history of the cosmos anyway, but their theories would inevitably  fall victim to discovery of anomalous data.   Religious thinkers from other traditions would almost certainly seize on the opportunity provided by the inability of scientists to explain major features of the cosmos by providing their own explanations.  These explanations would almost certainly be incorrect and if subject to falsification by scientific methods would also fall victim eventually to anomalous data.  The position of those holding to God's revelation,  principled agnosticism, might gain strength over time.

It is possible that the scientific community would eventually arrive at a consensus that it had "hit a wall" regarding an explanation of the history of the cosmos.  Further theoretical progress would not be possible unless the community found a way to gather new types of evidence that can help it penetrate the mysteries it keeps bumping up against.  In the past scientists have been able to overcome this type of barrer by developing new instruments and methods.  That could happen in this case too, but under all but the most optimistic and unlikely outcomes it would only put off the day of reckoning.   For example, if humans found a way to break out of our cosmos and survive in the outerverse the previously unaccountable mysteries could come under examination.

But the far more likely outcome is what we are interested in here.  Faced with apparently insurmountable roadblocks to theoretical progress using standard scientific methods and procedures, what alternatives, if any, would the community consider?  We have already ruled out revelation.  Deductive reasoning has already been thoroughly discredited. 

What non-empirical methods are left?   Transcendantal argument?  "Unless we assume the existence of an outerverse we have no way to account for the phenomena we encounter in our cosmos."   This is really just a sloppy, vague, and untestable variant of standard scientific method.  When the argument is provided with a mathematical form and proposes specific tests using detectable, observable, measurable data that can confirm/disconfirm it, it would get promoted to a genuine theory and no longer be an alternative to normal scientific procedure.  Otherwise, it is a cheap substitute and scientists would be likely to say, "Interesting but unhelpful." 

Would there be a divide between "believing" and "unbelieving" scientists over this matter?   I can't come up with a reason why there would be.  What would either party gain by opening up science to non-empirical methods?  Do they offer a new kind of evidence that can stand up to critical scrutiny, that will inspire greater confidence in the theory by friend and foe alike?  No.  This would be enough for the vast majority of the community, "believer" and "unbeliever" alike, to reject them.

In the end, the scientific community would be more likely to insist that their theories stay tied to detectable, obvervable, measurable data and live with incomplete theories than they would be to change their requirements so that they could have complete explanation.   In the next post I would like to begin an examination of approaches to "methodological naturalism" in the Hebrew Bible and the New Testament.

Thursday, July 15, 2010

Further Reflections on Methodological Naturalism as Detectability

In a previous post I suggested that sometimes theists and "naturalists" may, in the heat of their disputes, confuse the issue of "methodological naturalism."  Anti-theistic philosophers and scientists share with theists a tendency to fixate on  their differences over teleology, "supernatural" intelligence or design in nature and may more or less often drag this issue into disputes over "methodological naturalism."  In fact there is no necessary connection  between "methodological naturalism" and any position on the existence of design in nature, for example.  "Methodological naturalism" is short-hand for insisting that scientific explanation confine itself to entities, processes, and events that can be detected by human beings, whether by direct sense experience or the use of instruments.

I think it is possible that some of the folks with whom I've disputed over this matter made this confusion.     For that reason I plan to abandon the term "methodological naturalism" for the duration of this series of posts and talk about "detectability" and scientific method.

In her article on methodological naturalism Barbara Forrest quotes Arthur Strahler to the effect that introducing a single "supernatural" -- i.e, undetectable -- cause anywhere in a chain of causes making up a scientific explanation invalidates the entire explanation.  Is this single invocation of an undetectable cause really so devastating?  I think Strahler overstates the case, but qualified properly it still raises a legitimate issue.

Let's say, for example, I am sitting down to eat a stack of blueberry pancakes for breakfast and I suddenly have an overwhelming sense of deja vu.  In my mind I am carried back to a memory of breakfast at a summer camp.  I was at summer camp last year, but the camp in my memory is not the same as the one I actually attended.  In fact, the location and the people sitting next to me don't resemble those at any summer camp I can remember attending.  Joe psychologist explains to me that the smell and look of the pancakes triggered associations in my mind.  The unrecognized location and visitors come to me by way of reincarnation.  In a previous life I attended that very summer camp and sat with those very people for a breakfast of blueberry pancakes.

Strictly speaking, the event of reincarnation is unobservable.   Sure, I can detect one of its supposed effects, the apparent memory of people and a place I otherwise don't recall, but the event itself is beyond reach.   Nevertheless, for argument's sake let's say Joe psychologist is right.  Furthermore, let's say  that my experiences all conform to his diagnosis.  I have repeated, striking experiences of deja vu, vivid dreams or nightmares, and always they include things that I have no memory of otherwise.  Furthermore, let's say that all unrecognized details of these experiences can be traced to things experienced by the person I supposedly once was.

I can invoke several other undetectable causes to account for all the same phenomena, such as the presence of a demon who stole all these experiences from the other person and is planting them in my mind, or the existence of a post-mortem psychological memory "field" that radiates from the deceased person to another individual who can receive the memories on that "frequency."    To the extent that these and other possible undetectable causes can account for all my experiences, past, present, and future, it doesn't really matter which one is correct.  In fact, any and none of them could be correct, which means that Joe psychologist's explanation really amounts to "I don't know."  This is one way Strahler may be thinking that invoking an undetectable cause invalidates scientific explanation.  It amounts to an admission of ignorance. 

I've already hinted at a qualification to Strahler's argument.  Even undetectable causes can generate specific predictions about future observable events.  Assuming the predictions following from various undetectable causes differ, one can falsify one or more of them if predictions fail.  In this case the invocation of a specific undetectable cause is at least an implicit rejection of its undetectable competitors.     

On the other hand, Strahler is on to something genuinely pernicious about invoking undetectable causes.  Now that I have become convinced that Joe Psychologist is right, I am no longer driven to figure out where these apparent memories came from.  I ascribe them all to my past life.  With my critical faculties lulled to sleep I miss the clues, subtle or not,  leading to a simple explanation that does not invoke an undetectable cause.   To this someone will say, "Hey, the problem with reincarnation is not that it is undetectable, but that it is wrong!"

It certainly is wrong.  Strahler may be drawing from historical experience here.  Humans have often resorted to undetectable causes to explain the world, and in most cases their explanations were just plain wrong.  But the invocation of the undetectable cause, wrapped in religious language and delivered with divine authority, became an obligation.  To question it became a sin. 

The appeal to an undetectable cause often discourages scientific research at precisely the point we would otherwise expect it to intensify.   It is routine for scientists to encounter anomalies in their data, focus research on the anomalous situation, uncover new data, and use it to craft theories that resolve the anomaly.   Reincarnation may explain my apparent memories, but further research may uncover more mundane causes and obviate the need to invoke reincarnation.  This has happened often enough in the past, as Forrest points out in her articles.  And it is this historical experience that grounds statements like those of Lewontin.  The "a priori" refusal to allow undetectable causes in scientific explanation is really a lesson learned.  Religions make many claims about undetectable causes for what appear to be anomalous events in the world but on further investigation it turns out that detectable causes fully account for them.
Does this mean that a scientist who happens to be a believer should pursue research that could lead to the falsification of his beliefs?   Damn right!  I will work this out in detail when I get back to my series on Eve as a scientist in the garden of Eden.

Saturday, July 10, 2010

The Case of Rupert Sheldrake Illustrates how Some Evangelicals May Misunderstand Methodological Naturalism

Lately it seems I haven't been able to stick with multi-post projects on this blog.  This post may appear to be part of that pattern, but I'm hoping to link it in with my planned thought experiment on Eve in paradise.  The stimulus for this post was a series of exchanges I had with Steve Hays et. al. on Triablogue here and here regarding methodological naturalism in scientific investigation.

When I asked him for examples of scientific investigation that avoided the pitfall of methodological naturalism, he directed me to Rupert Sheldrake and Stephen Braude.  I took some time to learn about Rupert Sheldrake.

Reflection on Sheldrake's views and the controversy they have generated clarified a few things about the accusations some theists make against methodological naturalism in science.  It led me to do a little more digging and I found two  helpful articles by Barbara Forrest here and here along with some criticism. I would like to review this issue a bit again using some of the insights from the Forrest articles and criticisms before I tackle Sheldrake.

In the course of the discussion on Triablogue, one of the critics of methodological naturalism suggested that the real issue is not so much "natural" vs. "supernatural" as "designed" vs. "non-designed."   Some proponents of methodological naturalism might seem to agree.  For example, consider these words of Paul Kurtz:  "First, naturalism is committed to a methodological principle within the context of scientific inquiry; i.e., all hypotheses and events are to be explained and tested by reference to natural causes and events. To introduce a supernatural or transcendental cause within science is to depart from naturalistic explanations. On this ground, to invoke an intelligent designer or creator is inadmissible. . . ."  Or how about this from Steven Schafersman: “except for humans, philosophical naturalists understand nature to be fundamentally mindless and purposeless. . . ."  On this interpretation methodological naturalism betrays its precommitment to metaphysical naturalism by excluding in advance intelligent purpose, plan, or design from its theories about the operation of the world, except in the case of human endeavors -- and those of intelligent "aliens," assuming any are discovered.  "Supernatural" is the naturalist's code word for "deriving from a transcendant intelligence, especially God."  Another way to put it is that on this interpretation disputes over methodological naturalism largely resolve to the dispute over teleology.

Even after all this I think it is a fundamental mistake to confuse methodological naturalism with a resistance to teleology.  There is no necessary connection between them.  In fact, "naturalism" or "materialism" does not refer to an entity's lack of consciousness or purposive activity.  Instead, methodological naturalism is primarily a commitment to incorporating only what can be experienced by human beings into scientific explanation.  Forrest quotes Arthur Strahler to this effect:  "[S]upernatural forces, if they can be said to exist, cannot be observed, measured, or recorded [emphasis mine] by the procedures of science--that's simply what the word 'supernatural' means.  There can be no limit to the kinds and shapes of supernatural forces and forms the human mind is capable of conjuring up 'from nowhere.' Scientists therefore have no alternative but to ignore the claims of the existence of supernatural forces and causes."  Only if we can detect and measure a phenomenon can it play a role in scientific explanation. 

I believe this is the proper place to bring into play Richard Lewontin's notorious statement in his review of Carl Sagan's The Demon-Haunted World that scientists embrace materialism "a priori" and cannot let a Divine foot in the door.   He sums up as follows: "To appeal to an omnipotent deity is to allow that at any moment the regularities of nature may be ruptured, that miracles may happen." It sounds like Lewontin is complaining about God's apparently arbitrary intentionality.  But that is not the issue.  He hints at the real issue in the title to his review:  "Billions and Billions of Demons."  Demons, angels, fairies, aliens with superior technology, God himself all have the capacity, according to at least some of their devoted followers, to become totally undetectable at will.  They inhabit the world of human experience and an alternate world totally beyond human reach and can step in and out of that other world any time they want.   This convenient ability allows those who believe in them to dismiss scepticism because the inhabitant of that other world simply decided to hide herself from the sceptical inquirer.  This was a common tactic in the witch trials of the 16 and 17th centuries.  With that kind of argument you can spawn an arbitrary number of theories of arbitrary complexity populated by an arbitrary number of undetectable entities -- "billions" of demons.   Now replace the personal entity with a mystical experience such as reincarnation.   If a believer in reincarnation defends the belief with the same type of argument -- the evidence is simply hidden from the sceptical inquirer -- he runs afoul of methodological naturalism without making any teleological claims.   

This brings me to Rupert Sheldrake.  Sheldrake's theory of morphic resonance raises a lot of questions in my mind.  Perhaps reading his books would answer some of them.  For now, though, I just want to point out the following:
1.  The two critics of Sheldrake I heard from (Lewis Wolpert and P. Z. Myers) both stated clearly that they object because Sheldrake can present no good evidence for his theory.  Neither of them complained about Sheldrake's claims of divine intervention or supernatural powers, because he makes none.
2.  Sheldrake himself insists that he is working on accumulating evidence to back up his theories.  The experiments posted on his website are meant to gather data that will support the idea that there is some kind of collective "memory" or "field" shared by all members of a given species and that information gathered by one member is shared with all via transmission through this "field."  Steve Hays pointed to Sheldrake's experiments as an example of how one could do science without relying on methodological naturalism.  In fact, Sheldrake's experiments do no such thing.  His theory may be wacko mystical nonsense, as Wolpert claims, but his intent is to accumulate enough data to build an explanatory framework that conforms to the requirements of methodological naturalism.  He appears to believe, for instance, that we will eventually be able to detect and measure "morphic fields" directly.

Detectability is the real issue in methodological naturalism.  But detectability is a tricky issue on its own.  I reserve further discussion for another post.