The most important thing about science

The world of ideas in which we live

Consider the following statements:

• If anything bad can happen, it will.  For example, the GPS will always tell me that my destination will be on the left (i.e., across traffic; in England, Australia, or Malta, this would instead be on the right).
• Nice guys finish last.
• Good things happen to good people.
• Everything that has happened to me in my life – everything good, everything bad – has happened for a reason.  It is part of a divine plan.
• My car won’t start because the alternator has failed.
• Heavier objects fall faster than lighter objects.
• The universe was created in six days.  Living organisms were created on the fifth and sixth days and belong to a limited number of immutable species.
• The world is flat and exists on the back of a turtle.
• The Sun and all of the rest of the celestial bodies revolve around the Earth.
• The universe is almost 14 billion years old, the Earth orbits the Sun and is approximately 4.5 billion years old, life on earth began sometime before approximately 3.5 billion years, and organic species have been evolving and going extinct ever since.
• The evolution of life on Earth is explainable in terms of natural selection and only natural selection.
• Natural selection happens when an organism changes its traits to be better adapted to its environment, then passes these traits on to its offspring.
• Those individuals that survive longer are more evolutionarily successful.
• Those individuals that have the most offspring are more evolutionarily successful.
• The evolution of intelligent lifeforms is the end goal of evolution.
• Natural selection is the process by which species become better adapted to their environments; when confronted with environmental change, all species must change to adapt to new conditions.
• If I pour the coconut milk slowly into the blender, I will get more out of the carton than if I pour it quickly.
• If I play the Powerball numbers that have been drawn most frequently in the past, I will increase my odds of winning.
• Denali is 20,320 ft./6194 m tall.
• The distance between a lightning strike and its observer is approximately s/7 miles, where s is the number of seconds between the observation of the lightning and of its thunder peal.

To be human is to live in a world of ideas like these, ideas that assert something about the way the world looks and works, and why it works in one particular way rather than some other.  I say "ideas like these" because I do not intend this list to be exhaustive, nor to imply that every individual who has ever been born has shared all of the same beliefs.  The staggering plurality of beliefs held by the world's current population (>7 billion and growing), which underlie many of the bitter conflicts that are all too familiar to us, are proof enough that one could never realistically hope to inventory them all.  If we expand our scope to include every belief ever held in the 200,000 years since the dawn of the human mind as we know it, with its insatiable thirst for knowledge, then we have to add more than 75 billion human minds (though admittedly, many of these people did not survive infancy).  That's a lot of minds to change, and a lot of time to change our minds.

And change them we have.  I change my mind on a daily basis, though about some things more than others.  I inherited some of my current ideas from my parents, some from my teachers, some from my peers, some from strangers (books, magazines, radio, television, blogs), and many directly from my own experience of the world.  I have been accumulating beliefs for decades, but I also have not hesitated to abandon some of them whenever I have seen fit, sometimes in favor of better beliefs, but sometimes just because some have seemed so bad upon further examination.  I still have plenty of beliefs left, though.  I have so many of them, in fact, that I don't always recognize when I am holding contradictory ones. 

I hold my beliefs at every level of consciousness, from the "important" ideas that I rehearse with regularity, to the intuitions that go unspoken most of the time and that I am only vaguely aware of even when I try to think about them, to the matters that my body knows at such a basic level (motor intuitions) that I am not even consciously aware of them, nor could I ever be.

The ideas I have are about everything.  I have ideas about how old our universe, our planet, and life on our planet are; ideas about the general processes that led up to their present state and the specific changes that they have undergone along the way (quite vague when it comes to our universe and our solar system, admittedly); and ideas about how they work in the present.  I have ideas about the daily mechanics of life – about gravity, resistance, speed, inertia, temperature, etc. I have lots of ideas about human nature and about the way that human nature affects the ways that our social, economic, and political lives play out.  I have all kinds of ideas about health, disease, birth, and death.  I have ideas, and I have ideas, and I have ideas.  My cup runneth over with ideas.

Some of my ideas are quite important to me, though not all for the same reason: some are ideas that I take for granted when I make my practical daily decisions, some are the ones I have dedicated my life as a scientist to, some don’t affect me but I can’t imagine them being untrue, and some even bring me considerable displeasure, yet I believe them just the same.  Some I believe despite the fact they are demonstrably absurd (early in the morning, that thing about pouring coconut milk slowly from a finite cartonful is far more true for me than I care to admit), while others are so basic to my grasp on reality that I might have called them “self-evident” were I living in the 18th century (though it’s just not cool to describe ideas that way anymore).  The ideas that I have abandoned along the way have also varied in their importance to me.  With little more than a raised or furrowed eyebrow, I have cleared up misunderstandings about tax code, postal policy, and why the fridge has been making that funny sound (the condensor coil on the back of the fridge had been too close to the wall behind it, nothing more serious).  But others have shaken my sense of order and direction to their foundations.

If we assume that the thirst for understanding is a pervasive feature of human nature (and I do), this makes the business of abandoning beliefs puzzling, all the more when we further observe that we hold on to some longer than others.  Why can't or shouldn't we just keep the ones we already have and be done with it?  I assume that it is not for no reason, and this brings me to my main point.  It is such an important point that, if you take nothing else away from this blog, if this post is the first and last time you'll ever read it, I will be satisfied that at least it was this one.

The single most important thing about science

The single most important thing about science, the thing that sets it apart from all other ways of knowing, is that it focuses on scrutinizing beliefs, using observation of the world that these beliefs purport to understand to do so.  When I say that science is about building better beliefs, such scrutiny is at the very center of this enterprise.  Granted, building better beliefs must involve more than just scrutinizing them, particularly because scrutiny often leads to their abandonment.  To think otherwise is to be comfortable with the idea of eventually living a belief-free life, and that quite simply won’t do.  So obviously, there also has to be something additive to science, to offset the part of it that is subtractive.

Speaking metaphorically, we might think of the labor of science as being divided up into two complementary divisions: quality control, and research and development (R&D for short).  The people in quality control are charged with the task of inspecting the product, whether this be newly manufactured beliefs (i.e., the guesses or hunches that I mentioned previously) or the ones that have been on the market for a while.  On the other hand, the people in R&D are charged with coming up with new products (again, guesses), whether these be invented to explain newly discovered phenomena, or long-standing mysteries, or even to replace older, widely accepted explanations.

While it is true that these two metaphorical divisions complement each other, the quality control aspect is far more important to science.  I do not mean that scrutinizing ideas is somehow more important than coming up with new ideas, nor that coming up with new ideas is better than keeping old ones around.  I simply mean that, if we are trying to determine what sets science apart from other activities, it is the premium that science puts on scrutiny, and scrutiny based on observation of the world in particular.  Science insists that our understandings of the world be constrained by our observation of it, and this insistence is so central to it that no enterprise can rightly call itself ‘science’ if it does not take observational scrutiny seriously, if its practitioners are not sincerely open to endangering and potentially defeating ideas.  On the other hand, the activities of the R&D division are hardly unique to science.  On the contrary, there are many ways of coming into possession of beliefs, whether these be the time-honored beliefs that are passed down from generation to generation or those that have been newly invented by the revolutionaries of the avante-garde.  So, if science is to be set apart on the basis of its deeds, these will be the work of the folks over in quality control.

When I say "observation of the world that these beliefs purport to understand," what I mean by 'observation'  is "descriptions of the world, gained either through direct sensory perception or through the use of various measurement instruments (scales, measuring tapes, satellite arrays, mass spectrometers, etc.)."  Why?  Because if our understandings about the world are successful, if they actually help us to explain the phenomena we perceive, then what we actually perceive, including what we have perceived in the past and what we can perceive in the future, ought to be consistent with such understandings.  It is this risk of poor fit that makes observation so dangerous for ideas, that makes ideas so vulnerable to the risk of defeat:

In this framework, we give a new name to the idea – ‘hypothesis’ – but only if we are able to identify one or more kinds of observation that would allow us to test it.  Many scientists would also insist that we actually be able to make such observation, and I suppose that effectively, this is true; if we were never able to make one or another of the dangerous observations that we have identified, then the moment of peril would remain out of our reach.  Yet even so, it is a big step simply to be able to admit that one’s ideas are potentially vulnerable to observational scrutiny, and I consider any individual who is willing to hold their beliefs up to observational scrutiny, or who is at least dedicated to understanding the work of professionally trained scientists who do, to be a participant in the scientific enterprise.  This openness is very different from the person who is so committed to their understandings that, as a matter of principle, they refuse to let the world say otherwise.  It is also very different from the person who contends that his or her beliefs are informed by observational evidence but who chooses to emphasize only the evidence that is consistent with those beliefs.

Vexingly, it is also very different from some of the particularly smug scientists over in R&D, who are so enamored by the elegance of the ideas they have brainstormed that they find the world, not their ideas, to be in contempt when the fit between the two proves to be poor:

"On occasion, Einstein not only ignored the observational and experimental facts, but he even denied them.  Asked what would be his reaction to observation evidence against the bending of light predicted by his theory of general relativity, he answered, 'Then I would feel sorry for the good Lord.  The theory is correct anyway'"  (Hans C. Ohanian, Einstein’s Mistakes: The Human Failings of Genius, p. 5).

The tension that exists between the quality-control scientists and their colleagues in R&D is well-captured (albeit exaggerated for comedic effect) in the antagonism between The Big Bang Theory’s Dr. Sheldon Cooper (a theoretical physicist, i.e., an R&D guy) and Dr. Leonard Hofstadter (the experimental physicist, i.e., a quality control guy).  More realistically, theoretical and experimental physics are complementary operations, no less than any other scientific discipline's quality control and R&D divisions (see here for a real theoretical physicist's perspective on the matter).  I will have much more to say about the operations of the R&D division in my next post.

What is the outcome of such scrutiny?  Possibly, the failure of the idea under investigation.  The quality control operation, as I mentioned, is a subtractive one, though to be more accurate, it is a non-additive one: at the end of the operation, no new idea has been added; one has either been retained or eliminated.  Not that we should think that the dismissal of ideas is a bad thing; if we have rid ourselves of an idea, it is only that we have gotten rid of a demonstrably bad one, "vanquished the impossible" as Carl Sagan has said (though "vanquished the improbable" is a better way of putting it).  And the virtue of this elimination is amplified by the fact that it also clears up space for better beliefs.

But what if the scrutiny "fails" to defeat the idea?  On the one hand, ideas that don't fall to scrutiny start to look pretty good, all the more so the longer they stand up to scrutiny.  This is how scientists transform a practice focused on scrutiny to one capable of lending support.  On the other hand, such success should not be confused with proof.  "The hypothesis is consist with the evidence" is too weak a support to constitute an incontestable proof, so our uncertainty about the idea necessarily lingers.  Most scientists will be the first to admit that they are not in the business of proving anything, and those who do choose to use the word 'proof' either mean something different by it than 'incontestable proof' or they misunderstand the limits of their own methodology, or worse still, they are con men charading as scientists.  Most scientists would instead say that any hypothesis that has withstood scrutiny up to the present should be provisionally included as an item of standing knowledge (referring back to my earlier assertion that current definitions of knowledge are more inclusive than the stringently high standards imposed by Platonic epistemology), ever vulnerable to future scrutiny and to the risk of defeat that such entails.  By implication, we can say that it is possible to know something that is wrong, and by extension that we can know something today because it has been well-supported up to the present that we might know no longer tomorrow, once its support has been pulled out from under it.

Scrutiny based on observation is an imperfect mode of evaluation for another reason, and scientists have never claimed otherwise, but it bears repeating because it is not well enough recognized by the public at large: while remaining open to the possibility of scrutinizing our beliefs makes us all scientists of a sort, we cannot all be great or even good scientists, because not all kinds of observation are equally reliable (accurate or precise); the best ones, I am sorry to say, are technically demanding and expensive to operate.  Indeed, many of the intuitions we live by are informed by our five or so senses, yet these are considerably less reliable than we sometimes like to believe, or else we have subjected our intuitions to only the most casual of scrutiny using them.  As a wearer of glasses since age 8, I am keenly aware of the limitations of my own sense of vision, and even if I trusted it, I doubt that I ever would have come to the conclusion on my own that heavier and lighter objects fall at the same speed, all else being equal.  So, in order to improve the quality of the observations we use, we also have to scrutinize the quality of the observational methods we use to generate them.  To this end, some scientists dedicate their entire careers to identifying better and worse modes of observation.  Unfortunately, due caution regarding the limitations of science's observational methods is not always clearly conveyed to the public in the popular media.  For example, while media coverage of a recent re-measurement of the height of Denali suggests that it has shrunk by 83 ft. (25 m) over the last six decades (see also here), little attention has been paid to the possibility that this change is the result of imprecision inherent in the various methods used to estimate the mountain’s height.

In thinking in such great detail about the quality dimension of science, there are two mistakes that we should avoid making at all costs.  The first is lumping together untested ideas and bad ones.  Again, what makes an idea bad is its poor fit with the world it is supposed to explain, and this quite simply cannot be known until the idea has actually been tested.  Granted, untested ideas are a liability because they may be wrong, but this liability is counterbalanced to the degree that some of them will also eventually turn out to be good guesses.  In fact, if untested ideas were automatically deemed bad by virtue of their untestedness, this would disallow the possibility of ever having a good idea, because all of the good ideas that we have ever held, that we now hold, or that we might ever hold in the future began their lives as untested ideas, too.

The second mistake that we should avoid is the “Gee whiz, Mr. Science!” reflex, by which I mean the inclination to offhandedly dismiss any research that seems to do little more than support ideas that we already think we know.  At first site, such research does seem wasteful; if we knew it already, why not use the time, effort, and funding to explore new horizons instead?  Yet, the appearance of wastefulness is an illusion, one that persists right up until somebody's research reveals the falsehood of some confidently held, previously unexamined belief.  A more appropriate response to research that has demonstrated the "obvious" truths would be one of appreciation, because we no longer have to accept them with such blind faith.

But why abandon old beliefs, especially in favor of new guesses?

Okay, so science is in the business of snooping out ideas that seem to fit poorly with the world they are supposed to illuminate and of recommending these ones for termination, but again, the question is not so much whether we can scrutinize or refute beliefs but why we would ever want to do so.  Granted, for many the answer is no more complicated than a desire to possess only the highest-quality beliefs available; their appetite for belief is conditional on the quality, not about the quantity.  For others, however, the stakes in giving up beliefs are high enough – the sense of order that their beliefs afford them is salient enough – that the gamble entailed by scientific scrutiny is just too dangerous to accept.

The beliefs that afford a sense of purpose and/or hope for the future are an especially sore subject, and yet science's most outspoken advocates have not been particularly bothered by this matter.  Carl Sagan is well-known for his insistence that there are no ideas important enough to elude its scrutiny.  On the contrary, he said, science is intent on the pursuit of truth no matter where it leads, in other words no matter how psychologically disconcerting its revelations might be.  Likewise, Richard Feynman declared that he is less afraid of doubt than of the prospect of accepting beliefs that might be wrong.  In "vanquishing the impossible" (Sagan again), science is no less likely to scrutinize the hope-filled beliefs, nor to leave them in its wake if found wanting, than any other.

The scientific refusal to compromise on what beliefs we are willing to subject to scrutiny is an understandably unsettling prospect for many, but a few words should be said in rejection of the occasional claim that we are disporportionately invested, therefore persecutoral, in challenging the sacred.  I won't deny that this is a possibility on a small scale; scientists are humans, and this means that individual scientists are capable of dedicating their life's work to the assault on the sacred.  But we are not talking about potential, and for the most part scientists are no more mean-spirited than anyone else.  On the contrary, many of us care a great deal for the well-being of our fellow human beings, and on occasion scientists distress even ourselves by undermining our own sacred beliefs.  Even conceding that a good many sacred beliefs concern cosmic, planetary, and biological origins and that a good many scientists are dedicated to the study of such topics, it cannot be stressed enough that the overwhelming majority of such scientists do so out of sincere curiosity regarding such matters, not because they are spitefully compelled to inflict life-altering emotional trauma upon others, even if it does happen as a side effect.

It would also be wrong to limit our understanding of "important beliefs" to the sacred tenets of religious faith.  If the importance of a belief is measured by the trauma we experience upon its defeat, then it is easy to demonstrate that a good many of them have nothing at all to do with what we would consider sacred, for example this amusing anecdote from "The West Wing" (a fictional one, but one that probably feels familiar to many):

(In fact, no cartographic projection is a perfectly accurate representation of the Earth's or any other celestial body's surface, because all of them are equally guilty of the crime of representing spheroid surfaces as flat ones.  My favorite mind-blowing projections are the azimuthal equidistant projections, like this one centered on Hana, Hawai'i, in which Africa is wrapped around the entire perimeter of the map.)

http://www.wm7d.net/azproj.shtml

I am probably not alone in the fact that, as an 8th grader, I was baffled to learn that heavier and lighter objects fall at the same speed as each other, all else being equal.  (Conversely, I am amused by the general indifference of pretty much everyone, mountaineers included, to the demotion of Denali's stature.)

Alternatively, if the importance of a belief is measured by its relevance to our daily lives and decisions, then once again we find many more examples that fall outside the realm of the sacred.  Not surprisingly, the scientific scrutiny of these beliefs is frequently met with much the same zealous fervor and vehement defense as when scientists challenge the sacred – the recent food safety firestorm set off by an NPR story discouraging people from washing their chickens as a matter of good hygiene is an excellent example, with similar revolts lurking just below the surface regarding male circumcision, the safe cooking of pork, and the mixing of hot water and bleach – and yet accusations of persecution are peculiarly absent in such cases (or I have missed them).  Similar revolts are waiting in the wings.

However disinclined we may be to seeing our cherished beliefs defeated, the stakes in continuing to hold onto them may be high, as well.  As a consequence, we often find ourselves confronted by a dilemma of conflicting urgencies: to risk the psychological trauma of being left without previous hope or direction, or to accept the high physical, emotional, or financial tolls resulting from our continued acceptance of the false ones.  As we might expect, we encounter a mixed response among religious communities when it comes to the scrutiny of sacred dogmas: some certainly experience the scrutiny as persecution, as previously suggested, while others will be too attuned to the perils of embracing untrue beliefs to continue to indulge them with unremittingly blind faith:

(See the Dalai Lama's op-ed piece in the New York Times for the original quote.)

Similarly, while the sense of persecution is considerably diminished when scientists challenge the unreligious beliefs we live by (for example, the belief that washing a raw chicken is "a safer thing to do" than not to), the reception has nonetheless been quite mixed.  I don't wash chicken when I cook it, but apparently some people are really quite bothered that people like me exist.

In any case, it is probably fair to say that we are drawn to the pursuit of better beliefs not because of the intrinsic value of having good ideas but because of the practical value of good ideas.  The folly of embracing a bad idea is perhaps never more clear than when you turn the key to your car and it fails to start.  Is something wrong with the ignition system, or is it a dead battery?  If it is a dead battery, is something wrong with the alternator, is the battery old, or did you just leave the car door ajar overnight?  You begin to panic because you are going to be late to work, and you start considering what to do next to solve your predicament.  Jumping to any conclusion at such an early juncture would be folly; you certainly wouldn't refuse to consider the possibility that it is an ignition problem just because that would be more serious than a dead battery, but you also wouldn't run out and buy a new alternator out of blind faith that it must be an alternator problem.  When it comes to automotive problems, there is no room for sacred truths.

What would you do instead?  If you are inclined to automotive mechanics, you might attempt to diagnose the problem yourself.  Conversely, if you are like me, you would be better served to defer to people with such inclination.  In either case, the most beneficial approach to identifying the problem – not just coming up with any idea about what is wrong with your car, but coming up with a good one – will involve much the same sort of observational scrutiny as scientists advocate:

The "scientific method" is familiar enough so that it can be used intuitively.  In fact, all people, not just scientists, use it regularly.  Just listen to Click and Clack, the mechanics on Car Talk on National Public Radio, as they try to figure out what is causing a 1987 Volvo station wagon to stall unexpectedly as its driver, Bill from Beford, Massachusetts, motors down the highway.  (John Alcock, Animal Behavior 6th ed., p. 11)

Nor are the practical matters that we might address in such a manner limited to the personal ones.  The way that we approach our local, national, and global energy, health and safety, transportation, shelter, and food needs are influenced to a significant degree by our beliefs about them, so we have good reason to employ the best methods of critical thought available to size up our beliefs, including observational scrutiny.

Finally, the blind acceptance of many beliefs provides a potential for our disempowerment, to be exploited by con artists and tyrants.  The textbook example of belief-based exploitation is the 'divine right of kings,' which asserts the divinity or at least divine election of rulers and which has consequently been used to legitimate such rule at least since the dawn of written language (and probably well before it), to the detriment of hundreds of generations of loyal subjects.  In Enlightenment political thought, this belief was eventually replaced by the competing 'self-evident truth' that "all men are created equal," but in the United States at least, the belief in such equality was not legally extended to former slaves or their offspring, to women, or to Native Americans for a century or more.  It has often been observed that knowledge (or, to be more accurate, belief) is power, and those who are able to control it often do so with self-promoting agendas in mind, almost always to the detriment of others.  As Neil Degrasse Tyson has pointed out, our ability to scrutinize such beliefs thus affords us a powerful means to guard against the disempowerment and exploitation that blind faith entails.