Weiye Loh's Library tagged Falsifiability →  View Popular, Search in Google

if you were to do a survey of all the published research on any given question, you’d likely get a skewed perspective; and it would be skewed away from the null hypothesis. Thus, on any given question, the real answer is probably closer to the null hypothesis than a thorough meta-analysis would show.

 

The inevitable result is that to cite research that most accurately reflects what we’ve truly learned, it’s necessary to cherrypick. An unsavory thought, to be sure.

The larger and better the study, the more likely it is to be accurate. Therefore, when we logically put all of this together, larger and better studies are more likely to show results nearer the null hypothesis.

An entirely unspecified date has the advantage of making their prediction immune to refutation.  The fact that the rapture hasn’t occurred will never prove that it won’t occur in the future. A sense that they will never be refuted may well increase the subjective certainty of those who believe in the rapture, but this does nothing to provide the good reasons needed for objective certainty. Camping, after the fact, himself moved toward making his prediction unrefutable, saying that May 21 had been an “invisible judgment day,” a spiritual rather than a physical rapture. He kept to his prediction of a final, physical end of the world on October 21, 2011, but no doubt this prediction will also be open to reinterpretation.

Given their faith in the Bible, believers in the rapture do offer what they see as good reasons for their view as opposed to Camping’s.  They argue that the Bible clearly predicts a temporally unspecified rapture, whereas Camping’s specific date requires highly questionable numerological reasoning.  But many Christians—including many of the best Biblical scholars—do not believe that the Bible predicts a historical rapture.  Even those who accept the traditional doctrine of a Second Coming of Christ, preceding the end of the world, often reject the idea of a taking up of the saved into heaven, followed by a period of dreadful tribulations on Earth for those who are left behind.  Among believers themselves, a historical rapture is at best a highly controversial interpretation, not an objectively established certainty.

The case against Camping was this: His subjective certainty about the rapture required objectively good reasons to expect its occurrence; he provided no such reasons, so his claim was not worthy of belief.   Christians who believe in a temporally unspecified rapture agree with this argument.  But the same argument undermines their own belief in the rapture.  It’s not just that “no one knows the day and hour” of the rapture.  No one knows that it is going to happen at all.

Such environments eventually experience massive blowups, catching everyone off-guard and undoing years of stability or, in some cases, ending up far worse than they were in their initial volatile state. Indeed, the longer it takes for the blowup to occur, the worse the resulting harm in both economic and political systems.

Seeking to restrict variability seems to be good policy (who does not prefer stability to chaos?), so it is with very good intentions that policymakers unwittingly increase the risk of major blowups. And it is the same misperception of the properties of natural systems that led to both the economic crisis of 2007-8 and the current turmoil in the Arab world. The policy implications are identical: to make systems robust, all risks must be visible and out in the open -- fluctuat nec mergitur (it fluctuates but does not sink) goes the Latin saying.

Just as a robust economic system is one that encourages early failures (the concepts of "fail small" and "fail fast"), the U.S. government should stop supporting dictatorial regimes for the sake of pseudostability and instead allow political noise to rise to the surface. Making an economy robust in the face of business swings requires allowing risk to be visible; the same is true in politics.

Both the recent financial crisis and the current political crisis in the Middle East are grounded in the rise of complexity, interdependence, and unpredictability. Policymakers in the United Kingdom and the United States have long promoted policies aimed at eliminating fluctuation -- no more booms and busts in the economy, no more "Iranian surprises" in foreign policy. These policies have almost always produced undesirable outcomes. For example, the U.S. banking system became very fragile following a succession of progressively larger bailouts and government interventions, particularly after the 1983 rescue of major banks (ironically, by the same Reagan administration that trumpeted free markets). In the United States, promoting these bad policies has been a bipartisan effort throughout. Republicans have been good at fragilizing large corporations through bailouts, and Democrats have been good at fragilizing the government. At the same time, the financial system as a whole exhibited little volatility; it kept getting weaker while providing policymakers with the illusion of stability, illustrated most notably when Ben Bernanke, who was then a member of the Board of Governors of the U.S. Federal Reserve, declared the era of "the great moderation" in 2004.

Washington stabilized the market with bailouts and by allowing certain companies to grow "too big to fail." Because policymakers believed it was better to do something than to do nothing, they felt obligated to heal the economy rather than wait and see if it healed on its own.

The foreign policy equivalent is to support the incumbent no matter what. And just as banks took wild risks thanks to Greenspan's implicit insurance policy, client governments such as Hosni Mubarak's in Egypt for years engaged in overt plunder thanks to similarly reliable U.S. support.

Those who seek to prevent volatility on the grounds that any and all bumps in the road must be avoided paradoxically increase the probability that a tail risk will cause a major explosion.

In the realm of economics, price controls are designed to constrain volatility on the grounds that stable prices are a good thing. But although these controls might work in some rare situations, the long-term effect of any such system is an eventual and extremely costly blowup whose cleanup costs can far exceed the benefits accrued. The risks of a dictatorship, no matter how seemingly stable, are no different, in the long run, from those of an artificially controlled price.

Such attempts to institutionally engineer the world come in two types: those that conform to the world as it is and those that attempt to reform the world. The nature of humans, quite reasonably, is to intervene in an effort to alter their world and the outcomes it produces. But government interventions are laden with unintended -- and unforeseen -- consequences, particularly in complex systems, so humans must work with nature by tolerating systems that absorb human imperfections rather than seek to change them.

What is needed is a system that can prevent the harm done to citizens by the dishonesty of business elites; the limited competence of forecasters, economists, and statisticians; and the imperfections of regulation, not one that aims to eliminate these flaws. Humans must try to resist the illusion of control: just as foreign policy should be intelligence-proof (it should minimize its reliance on the competence of information-gathering organizations and the predictions of "experts" in what are inherently unpredictable domains), the economy should be regulator-proof, given that some regulations simply make the system itself more fragile. Due to the complexity of markets, intricate regulations simply serve to generate fees for lawyers and profits for sophisticated derivatives traders who can build complicated financial products that skirt those regulations.

The life of a turkey before Thanksgiving is illustrative: the turkey is fed for 1,000 days and every day seems to confirm that the farmer cares for it -- until the last day, when confidence is maximal. The "turkey problem" occurs when a naive analysis of stability is derived from the absence of past variations. Likewise, confidence in stability was maximal at the onset of the financial crisis in 2007.

The turkey problem for humans is the result of mistaking one environment for another. Humans simultaneously inhabit two systems: the linear and the complex. The linear domain is characterized by its predictability and the low degree of interaction among its components, which allows the use of mathematical methods that make forecasts reliable. In complex systems, there is an absence of visible causal links between the elements, masking a high degree of interdependence and extremely low predictability. Nonlinear elements are also present, such as those commonly known, and generally misunderstood, as "tipping points." Imagine someone who keeps adding sand to a sand pile without any visible consequence, until suddenly the entire pile crumbles. It would be foolish to blame the collapse on the last grain of sand rather than the structure of the pile, but that is what people do consistently, and that is the policy error.

Engineering, architecture, astronomy, most of physics, and much of common science are linear domains. The complex domain is the realm of the social world, epidemics, and economics. Crucially, the linear domain delivers mild variations without large shocks, whereas the complex domain delivers massive jumps and gaps. Complex systems are misunderstood, mostly because humans' sophistication, obtained over the history of human knowledge in the linear domain, does not transfer properly to the complex domain. Humans can predict a solar eclipse and the trajectory of a space vessel, but not the stock market or Egyptian political events. All man-made complex systems have commonalities and even universalities. Sadly, deceptive calm (followed by Black Swan surprises) seems to be one of those properties.

The system is responsible, not the components. But after the financial crisis of 2007-8, many people thought that predicting the subprime meltdown would have helped. It would not have, since it was a symptom of the crisis, not its underlying cause. Likewise, Obama's blaming "bad intelligence" for his administration's failure to predict the crisis in Egypt is symptomatic of both the misunderstanding of complex systems and the bad policies involved.

Obama's mistake illustrates the illusion of local causal chains -- that is, confusing catalysts for causes and assuming that one can know which catalyst will produce which effect. The final episode of the upheaval in Egypt was unpredictable for all observers, especially those involved. As such, blaming the CIA is as foolish as funding it to forecast such events. Governments are wasting billions of dollars on attempting to predict events that are produced by interdependent systems and are therefore not statistically understandable at the individual level.

Political and economic "tail events" are unpredictable, and their probabilities are not scientifically measurable. No matter how many dollars are spent on research, predicting revolutions is not the same as counting cards; humans will never be able to turn politics into the tractable randomness of blackjack.

Most explanations being offered for the current turmoil in the Middle East follow the "catalysts as causes" confusion. The riots in Tunisia and Egypt were initially attributed to rising commodity prices, not to stifling and unpopular dictatorships. But Bahrain and Libya are countries with high gdps that can afford to import grain and other commodities. Again, the focus is wrong even if the logic is comforting. It is the system and its fragility, not events, that must be studied -- what physicists call "percolation theory," in which the properties of the terrain are studied rather than those of a single element of the terrain.

When dealing with a system that is inherently unpredictable, what should be done? Differentiating between two types of countries is useful. In the first, changes in government do not lead to meaningful differences in political outcomes (since political tensions are out in the open). In the second type, changes in government lead to both drastic and deeply unpredictable changes.

Humans fear randomness -- a healthy ancestral trait inherited from a different environment. Whereas in the past, which was a more linear world, this trait enhanced fitness and increased chances of survival, it can have the reverse effect in today's complex world, making volatility take the shape of nasty Black Swans hiding behind deceptive periods of "great moderation." This is not to say that any and all volatility should be embraced. Insurance should not be banned, for example.

But alongside the "catalysts as causes" confusion sit two mental biases: the illusion of control and the action bias (the illusion that doing something is always better than doing nothing). This leads to the desire to impose man-made solutions

Variation is information. When there is no variation, there is no information. This explains the CIA's failure to predict the Egyptian revolution and, a generation before, the Iranian Revolution -- in both cases, the revolutionaries themselves did not have a clear idea of their relative strength with respect to the regime they were hoping to topple. So rather than subsidize and praise as a "force for stability" every tin-pot potentate on the planet, the U.S. government should encourage countries to let information flow upward through the transparency that comes with political agitation. It should not fear fluctuations per se, since allowing them to be in the open, as Italy and Lebanon both show in different ways, creates the stability of small jumps.

As Seneca wrote in De clementia, "Repeated punishment, while it crushes the hatred of a few, stirs the hatred of all . . . just as trees that have been trimmed throw out again countless branches." The imposition of peace through repeated punishment lies at the heart of many seemingly intractable conflicts, including the Israeli-Palestinian stalemate. Furthermore, dealing with seemingly reliable high-level officials rather than the people themselves prevents any peace treaty signed from being robust. The Romans were wise enough to know that only a free man under Roman law could be trusted to engage in a contract; by extension, only a free people can be trusted to abide by a treaty. Treaties that are negotiated with the consent of a broad swath of the populations on both sides of a conflict tend to survive. Just as no central bank is powerful enough to dictate stability, no superpower can be powerful enough to guarantee solid peace alone.

As Jean-Jacques Rousseau put it, "A little bit of agitation gives motivation to the soul, and what really makes the species prosper is not peace so much as freedom." With freedom comes some unpredictable fluctuation. This is one of life's packages: there is no freedom without noise -- and no stability without volatility.∂

So if syphilis causes AIDS, and not HIV, where is the evidence? As microbiologist and epidemiolist Tara Smith points out in her excellent blog, Margulis offers none. Instead, she says to the credulous and uncritical interviewer:

 

The idea that penicillin kills the cause of the disease is nuts. If you treat the painless chancre in the first few days of infection, you may stop the bacterium before the symbiosis develops, but if you really get syphilis, all you can do is live with the spirochete. The spirochete lives permanently as a symbiont in the patient. The infection cannot be killed because it becomes part of the patient’s genome and protein synthesis biochemistry. After syphilis establishes this symbiotic relationship with a person, it becomes dependent on human cells and is undetectable by any testing.

 

Great. Just what we need: an untestable hypothesis promoted by assertion and reputation, not something concrete that scientists could test (although most specialists in microbiology would say the evidence is clear that the HIV retrovirus, and not the spirochaete bacterium Treponema pallidum, is the true cause of AIDS).

On any new problem, of course, inductive inference is not as simple and certain as deduction, because it involves reaching out into the unknown. Steps 1 and 2 require intellectual inventions, which must be cleverly chosen so that hypothesis, experiment, outcome, and exclusion will be related in a rigorous syllogism; and the question of how to generate such inventions is one which has been extensively discussed elsewhere (2, 3). What the formal schema reminds us to do is to try to make these inventions, to take the next step, to proceed to the next fork, without dawdling or getting tied up in irrelevancies.

It is clear why this makes for rapid and powerful progress. For exploring the unknown, there is no faster method; this is the minimum sequence of steps. Any conclusion that is not an exclusion is insecure and must be rechecked. Any delay in recycling to the next set of hypotheses is only a delay. Strong inference, and the logical tree it generates, are to inductive reasoning what the syllogism is to deductive reasoning in that it offers a regular method for reaching firm inductive conclusions one after the other as rapidly as possible.

"But what is so novel about this?" someone will say. This is the method of science and always has been, why give it a special name? The reason is that many of us have almost forgotten it. Science is now an everyday business. Equipment, calculations, lectures become ends in themselves. How many of us write down our alternatives and crucial experiments every day, focusing on the exclusion of a hypothesis? We may write our scientific papers so that it looks as if we had steps 1, 2, and 3 in mind all along. But in between, we do busywork. We become "method- oriented" rather than "problem-oriented." We say we prefer to "feel our way" toward generalizations. We fail to teach our students how to sharpen up their inductive inferences. And we do not realize the added power that the regular and explicit use of alternative hypothesis and sharp exclusion could give us at every step of our research.

At the 1958 Conference on Biophysics, at Boulder, there was a dramatic confrontation between the two points of view. Leo Szilard said: "The problems of how enzymes are induced, of how proteins are synthesized, of how antibodies are formed, are closer to solution than is generally believed. If you do stupid experiments, and finish one a year, it can take 50 years. But if you stop doing experiments for a little while and think how proteins can possibly be synthesized, there are only about 5 different ways, not 50! And it will take only a few experiments to distinguish these."

  

One of the young men added: "It is essentially the old question: How small and elegant an experiment can you perform?"

  

These comments upset a number of those present. An electron microscopist said. "Gentlemen, this is off the track. This is philosophy of science."

  

Szilard retorted. "I was not quarreling with third-rate scientists: I was quarreling with first-rate scientists."

A distinguished cell biologist rose and said, "No two cells give the same properties. Biology is the science of heterogeneous systems." And he added privately. "You know there are scientists, and there are people in science who are just working with these over-simplified model systems - DNA chains and in vitro systems - who are not doing science at all. We need their auxiliary work: they build apparatus, they make minor studies, but they are not scientists."

  

To which Cy Levinthal replied: "Well, there are two kinds of biologists, those who are looking to see if there is one thing that can be understood and those who keep saying it is very complicated and that nothing can be understood. . . . You must study the simplest system you think has the properties you are interested in."

Any criticism or challenge to consider changing our methods strikes of course at all our ego-defenses. But in this case the analytical method offers the possibility of such great increases in effectiveness that it is unfortunate that it cannot be regarded more often as a challenge to learning rather than as challenge to combat. Many of the recent triumphs in molecular biology have in fact been achieved on just such "oversimplified model systems," very much along the analytical lines laid down in the 1958 discussion. They have not fallen to the kind of men who justify themselves by saying "No two cells are alike," regardless of how true that may ultimately be. The triumphs are in fact triumphs of a new way of thinking.

the emphasis on strong inference

is also partly due to the nature of the fields themselves. Biology, with its vast informational detail and complexity, is a "high-information" field, where years and decades can easily be wasted on the usual type of "low-information" observations or experiments if one does not think carefully in advance about what the most important and conclusive experiments would be. And in high-energy physics, both the "information flux" of particles from the new accelerators and the million-dollar costs of operation have forced a similar analytical approach. It pays to have a top-notch group debate every experiment ahead of time; and the habit spreads throughout the field.

Historically, I think, there have been two main contributions to the development of a satisfactory strong-inference method. The first is that of Francis Bacon (13). He wanted a "surer method" of "finding out nature" than either the logic-chopping or all-inclusive theories of the time or the laudable but crude attempts to make inductions "by simple enumeration." He did not merely urge experiments as some suppose, he showed the fruitfulness of interconnecting theory and experiment so that the one checked the other. Of the many inductive procedures he suggested, the most important, I think, was the conditional inductive tree, which proceeded from alternative hypothesis (possible "causes," as he calls them), through crucial experiments ("Instances of the Fingerpost"), to exclusion of some alternatives and adoption of what is left ("establishing axioms"). His Instances of the Fingerpost are explicitly at the forks in the logical tree, the term being borrowed "from the fingerposts which are set up where roads part, to indicate the several directions."

ere was a method that could separate off the empty theories!

  

Bacon, said the inductive method could be learned by anybody, just like learning to "draw a straighter line or more perfect circle . . . with the help of a ruler or a pair of compasses." "My way of discovering sciences goes far to level men's wit and leaves but little to individual excellence, because it performs everything by the surest rules and demonstrations." Even occasional mistakes would not be fatal. "Truth will sooner come out from error than from confusion."

Nevertheless there is a difficulty with this method. As Bacon emphasizes, it is necessary to make "exclusions." He says, "The induction which is to be available for the discovery and demonstration of sciences and arts, must analyze nature by proper rejections and exclusions, and then, after a sufficient number of negatives come to a conclusion on the affirmative instances." "[To man] it is granted only to proceed at first by negatives, and at last to end in affirmatives after exclusion has been exhausted."

  

Or, as the philosopher Karl Popper says today there is no such thing as proof in science - because some later alternative explanation may be as good or better - so that science advances only by disproofs. There is no point in making hypotheses that are not falsifiable because such hypotheses do not say anything, "it must be possible for all empirical scientific system to be refuted by experience" (14).

The difficulty is that disproof is a hard doctrine. If you have a hypothesis and I have another hypothesis, evidently one of them must be eliminated. The scientist seems to have no choice but to be either soft-headed or disputatious. Perhaps this is why so many tend to resist the strong analytical approach and why some great scientists are so disputatious.

Fortunately, it seems to me, this difficulty can be removed by the use of a second great intellectual invention, the "method of multiple hypotheses," which is what was needed to round out the Baconian scheme. This is a method that was put forward by T.C. Chamberlin (15), a geologist at Chicago at the turn of the century, who is best known for his contribution to the Chamberlain-Moulton hypothesis of the origin of the solar system.

Chamberlin says our trouble is that when we make a single hypothesis, we become attached to it.

  

"The moment one has offered an original explanation for a phenomenon which seems satisfactory, that moment affection for his intellectual child springs into existence, and as the explanation grows into a definite theory his parental affections cluster about his offspring and it grows more and more dear to him. . . . There springs up also unwittingly a pressing of the theory to make it fit the facts and a pressing of the facts to make them fit the theory..."

  

"To avoid this grave danger, the method of multiple working hypotheses is urged. It differs from the simple working hypothesis in that it distributes the effort and divides the affections. . . . Each hypothesis suggests its own criteria, its own method of proof, its own method of developing the truth, and if a group of hypotheses encompass the subject on all sides, the total outcome of means and of methods is full and rich."

The conflict and exclusion of alternatives that is necessary to sharp inductive inference has been all too often a conflict between men, each with his single Ruling Theory. But whenever each man begins to have multiple working hypotheses, it becomes purely a conflict between ideas. It becomes much easier then for each of us to aim every day at conclusive disproofs - at strong inference - without either reluctance or combativeness. In fact, when there are multiple hypotheses, which are not anyone's "personal property," and when there are crucial experiments to test them, the daily life in the laboratory takes on an interest and excitement it never had, and the students can hardly wait to get to work to see how the detective story will come out. It seems to me that this is the reason for the development of those distinctive habits of mind and the "complex thought" that Chamberlin described, the reason for the sharpness, the excitement, the zeal, the teamwork - yes, even international teamwork - in molecular biology and high- energy physics today. What else could be so effective?

Unfortunately, I think, there are other other areas of science today that are sick by comparison, because they have forgotten the necessity for alternative hypotheses and disproof. Each man has only one branch - or none - on the logical tree, and it twists at random without ever coming to the need for a crucial decision at any point. We can see from the external symptoms that there is something scientifically wrong. The Frozen Method, The Eternal Surveyor, The Never Finished, The Great Man With a Single Hypothcsis, The Little Club of Dependents, The Vendetta, The All-Encompassing Theory Which Can Never Be Falsified.

a "theory" of this sort is not a theory at all, because it does not exclude anything. It predicts everything, and therefore does not predict anything. It becomes simply a verbal formula which the graduate student repeats and believes because the professor has said it so often. This is not science, but faith; not theory, but theology. Whether it is hand-waving or number-waving, or equation-waving, a theory is not a theory unless it can be disproved. That is, unless it can be falsified by some possible experimental outcome.

the work methods of a number of scientists have been testimony to the power of strong inference. Is success not due in many cases to systematic use of Bacon's "surest rules and demonstrations" as much as to rare and unattainable intellectual power? Faraday's famous diary (16), or Fermi's notebooks (3, 17), show how these men believed in the effectiveness of daily steps in applying formal inductive methods to one problem after another.

Surveys, taxonomy, design of equipment, systematic measurements and tables, theoretical computations - all have their proper and honored place, provided they are parts of a chain of precise induction of how nature works. Unfortunately, all too often they become ends in themselves, mere time-serving from the point of view of real scientific advance, a hypertrophied methodology that justifies itself as a lore of respectability.

We speak piously of taking measurements and making small studies that will "add another brick to the temple of science." Most such bricks just lie around the brickyard (20). Tables of constraints have their place and value, but the study of one spectrum after another, if not frequently re-evaluated, may become a substitute for thinking, a sad waste of intelligence in a research laboratory, and a mistraining whose crippling effects may last a lifetime.

Beware of the man of one method or one instrument, either experimental or theoretical. He tends to become method-oriented rather than problem-oriented. The method-oriented man is shackled; the problem-oriented man is at least reaching freely toward that is most important. Strong inference redirects a man to problem-orientation, but it requires him to be willing repeatedly to put aside his last methods and teach himself new ones.

anyone who asks the question about scientific effectiveness will also conclude that much of the mathematizing in physics and chemistry today is irrelevant if not misleading.

  

The great value of mathematical formulation is that when an experiment agrees with a calculation to five decimal places, a great many alternative hypotheses are pretty well excluded (though the Bohr theory and the Schrödinger theory both predict exactly the same Rydberg constant!). But when the fit is only to two decimal places, or one, it may be a trap for the unwary; it may be no better than any rule-of-thumb extrapolation, and some other kind of qualitative exclusion might be more rigorous for testing the assumptions and more important to scientific understanding than the quantitative fit.

Today we preach that science is not science unless it is quantitative. We substitute correlations for causal studies, and physical equations for organic reasoning. Measurements and equations are supposed to sharpen thinking, but, in my observation, they more often tend to make the thinking noncausal and fuzzy. They tend to become the object of scientific manipulation instead of auxiliary tests of crucial inferences.

Many - perhaps most - of the great issues of science are qualitative, not quantitative, even in physics and chemistry. Equations and measurements are useful when and only when they are related to proof; but proof or disproof comes first and is in fact strongest when it is absolutely convincing without any quantitative measurement.

you can catch phenomena in a logical box or in a mathematical box. The logical box is coarse but strong. The mathematical box is fine-grained but flimsy. The mathematical box is a beautiful way of wrapping up a problem, but it will not hold the phenomena unless they have been caught in a logical box to begin with.

Of course it is easy - and all too common - for one scientist to call the others unscientific. My point is not that my particular conclusions here are necessarily correct, but that we have long needed some absolute standard of possible scientific effectiveness by which to measure how well we are succeeding in various areas - a standard that many could agree on and one that would be undistorted by the scientific pressures and fashions of the times and the vested interests and busywork that they develop. It is not public evaluation I am interested in so much as a private measure by which to compare one's own scientific performance with what it might be. I believe that strong inference provides this kind of standard of what the maximum possible scientific effectiveness could be - as well as a recipe for reaching it.

The strong-inference point of view is so resolutely critical of methods of work and values in science that any attempt to compare specific cases is likely to sound but smug and destructive. Mainly one should try to teach it by example and by exhorting to self-analysis and self-improvement only in general terms

one severe but useful private test - a touchstone of strong inference - that removes the necessity for third-person criticism, because it is a test that anyone can learn to carry with him for use as needed. It is our old friend the Baconian "exclusion," but I call it "The Question." Obviously it should be applied as much to one's own thinking as to others'. It consists of asking in your own mind, on hearing any scientific explanation or theory put forward, "But sir, what experiment could disprove your hypothesis?"; or, on hearing a scientific experiment described, "But sir, what hypothesis does your experiment disprove?"

It is not true that all science is equal; or that we cannot justly compare the effectiveness of scientists by any method other than a mutual-recommendation system. The man to watch, the man to put your money on, is not the man who wants to make "a survey" or a "more detailed study" but the man with the notebook, the man with the alternative hypotheses and the crucial experiments, the man who knows how to answer your Question of disproof and is already working on it.