When Experts Disagree, Do We Have Consensus?

Quote

When Expert Disagreement Supports the Consensus : Finnur Dellsén

Before some time a similar advices emerged: The era of expert failure by Arnold Kling, Why experts are usually wrong by David H. Freeman and Why the experts missed the crash by Phill Tetlock.

As I explained here, the expanding technology enabled us to observe hyperdimensional phenomena, which can be described from multiple observational perspectives. These perspectives may be even all perfectly logical and relevant - they're just incomplete and until someone hasn't deeper / more dimensional and complex understanding of reality, he isn't able to note their hidden connections. For laymen and even experts such a situation is indeed source of neverending confusion: for laymen because they simply have no time to study the particular problem deeper, but for experts this situation may get equally difficult, because their understanding remains limited to narrow perspective of their professional specialization.

In this situation the journalist persons, who keeps broader overview have an advantage over dedicated experts. Unfortunately the consensus ("Vox populi, vox Dei") may be equally bad adviser here, like the individual experts and it can be biased in its specific way.

Flaccus Alcuinus from York has said in AD 798: Nec audiendi qui solent dicere, Vox populi, vox Dei, quum tumultuositas vulgi semper insaniae proxima sit ("And those people should not be listened to who keep saying the voice of the people is the voice of God, since the riotousness of the crowd is always very close to madness)")

After all, which consensus exists at the case of overunity and cold fusion findings, global warming and similar stuffs?

In my personal experience, neither opinion of mainstream experts, neither wide consensus of these experts or laymen - but the consensus between distinct group of experts: the proponents of opposite dual perspective dual to mainstream gets most close to actual evolution of problem if future. This experience has its geometric analogy in formation of matter (substance which persists) inside the universe. In dense aether model the Universe is steady state but in stage of neverending exchange between matter and radiation. The matter gradually evaporates into a radiation (transverse and scalar waves, i.e. the dark matter), which condense somewhere else. The mechanism in which basic interaction work provide, that this condensation occurs at the sufficient distance from existing massive objects. It's because the dark matter particles are attracted to negative gradient of gravitational potential due to shielding mechanism of gravity. In this way, the dark matter (the dual paradigm of mainstream) concentrates along connection lines (shadows) of existing massive objects and it condenses there into a new generation of matter. The shielding supergravity mechanism has its analogy in the inference reasoning ("find the places, which most experts disagree in least way"). The new findings and ideas dual to mainstream paradigm concentrate in the same way between opponents of mainstream ideas.

Therefore the Bayesian reasoning it's still relevant - but it must be applied to particular group of experts, who are in opposition to mainstream - not all experts or even laymen population. It's because both mainstream experts, both laymen population are money and occupation driven - and as such biased - just in opposite ways (because laymen are actually who pays these experts). The actual truth is usually somewhere in midway between these two opinions, but it cannot be estimated naively like their average.

Quote from Zephir_AWT

Before some time a similar advices emerged: The era of expert failure by Arnold Kling, Why experts are usually wrong by David H. Freeman and Why the experts missed the crash by Phill Tetlock.

People who "miss the crash" and who make similar large mistakes are not experts. By definition, an expert is someone who is generally right. People who make large mistakes are wannabe experts. Citing them is a logical fallacy: "Fallacious Appeal to Authority," or "Questionable Authority."

Scientists who claim that cold fusion does not exist, or there is no evidence for it are wannabe experts, not real experts. Their assertions are factually wrong. In most cases, they are unaware of this. In a few cases, such as with Robert Park, they are lying through their teeth.

When you are outside of a field, and you yourself have no expert knowledge, it can be difficult for you to know who is a real expert and who only imagines himself to be an expert. You cannot judge. You have to assume that important people in the field who are called upon to write op-ed pieces in the New York Times, and who are appointed to high level government positions are actually experts. Often they are, but in some instances they are not.

This link describes the problem in detail:

http://www.nizkor.org/features…/appeal-to-authority.html

QUOTE:

Since this sort of reasoning is fallacious only when the person is not a legitimate authority in a particular context, it is necessary to provide some acceptable standards of assessment. The following standards are widely accepted:

1. The person has sufficient expertise in the subject matter in question.

   Claims made by a person who lacks the needed degree of expertise to make a reliable claim will, obviously, not be well supported. In contrast, claims made by a person with the needed degree of expertise will be supported by the person's reliability in the area.

   Determining whether or not a person has the needed degree of expertise can often be very difficult. In academic fields (such as philosophy, engineering, history, etc.), the person's formal education, academic performance, publications, membership in professional societies, papers presented, awards won and so forth can all be reliable indicators of expertise. Outside of academic fields, other standards will apply. . . .

Quote

By definition, an expert is someone who is generally right.

By its Latin origin expert is word borrowed from Ancient Greek’s second declension expertus, i.e. (well) tested, proved i.e. experienced (feminine experta, neuter expertum). The Niels Bohr's definition was exactly the opposite:

"An expert is a man who has made all the mistakes which can be made, in a narrow field." and he didn't mean it pejoratively at all.

The problem with experts is, they tend to disagree mutually once they're right whereas they tend to agree once they're all wrong. Getting liberals to agree is like herding cats ... The conservative experts aren't exploratory - instead of it, they tend to remain specialized to the subject of their expertise, thus being wrong soon or latter. Therefore, the disagreeing mutually is the first indicia of actual progress. Instead of it, the wide consensus is the first indicia of fundamental bias.

In dense aether model this behavior has its analogy in behavior of black holes (remnants of visible matter) and the dark matter (progenitor of visible matter). The black represent the past of Universe and they tend to evaporate soon or later. They remain cohesive, being formed with particles of positive space-time curvature. Instead of it, the particles of opposite space-time curvature are systematically expelled from them. Dark matter particles behave like sparse bubbles of space-time and they're repelling mutually at distance, thus remaining in diaspora. They're just attracted to existing observable matter, thus forming dark matter halo around massive galaxies and stars, which remains separated at distance. The experts to alternative physics behave similarly - they're expelled from mainstream and they're working in diaspora. The gradual increase of their concentration - not mutual agreement - is what indicates the progress and nearing technological transform.

Quote from Zephir_AWT

"An expert is a man who has made all the mistakes which can be made, in a narrow field." and he didn't mean it pejoratively at all.

Okay, but after you make all mistakes which can be made, and you are an expert, you should no longer made large mistakes. Not often, anyway. Such as not predicting the 2008 crash (the example from above). Of course everyone makes mistakes from time to time. But if you make significantly more than other experts, you are not an expert.

The people running IBM in the 1980s were not experts. Everyone assumed they were, but Paul Carroll of the Wall Street Journal described them:

"They debated new technologies, such as PCs, but the older executives who ran the company rarely dirtied their hands with the technologies enough to understand their potential and their problems. IBM had become like a music-publishing company run by deaf people."

- "Big Blues, the Unmaking of IBM")

Even indisputably world-class experts make small mistakes or careless mistakes from time to time. That doesn't count.

Quote

But if you make significantly more than other experts, you are not an expert.

OK, but the problem with this definition is, it's postdictive, not predictive (being tautological in fact). It doesn't enable you to judge people as experts before the final truth will turn out. And being relativistic, it enables to label experts in style "In the land of the blind, the one-eyed man is the king". In my understanding the expert is merely someone like the specialist - i.e. the person with interest/experience dedicated to a narrow area of expertise.

Quote from Zephir_AWT

Instead of it, the wide consensus is the first indicia of fundamental bias.

That would depend on what the consensus is. I think all modern doctors agree that the "germ theory" is correct -- that bacteria causes disease. Electricians agree that high resistance light bulbs can be wired in parallel. Mechanical engineers now agree that high pressure steam can be made reasonably safe, and gas turbine (jet) aircraft engines are a thing. (Real, that is.) Most people today do not think that the use of zippers in clothing leads to sexual abandon and the collapse of society. All biologists agree that evolution occurred, and all but a few neo-Lamarkians assume that Darwin's theory explains it.

Everything in that list was once controversial. These issues generated widespread opposition -- indeed, widespread hysteria in the case of zippers. There was no consensus at first. Only people like me who read a lot of history realize that. For example, regarding the "germ theory," today we all agree that pasteurizing milk saves lives and it reassures customers the product is safe. This was clear after 1870. But the milk producers in New York state refused to pasteurize until the government forced them to in 1917.

(My mother and other people from that era called it the "germ theory," but by that time they were being sarcastic. They knew bacteria causes disease. On the other hand, infant mortality caused by contaminated milk killed hundreds of thousands of babies in New York before 1917, so that part wasn't sarcastic.)

Quote from Zephir_AWT

OK, but the problem with this definition is, it's postdictive, not predictive (being tautological in fact). It doesn't enable you to judge people as experts before the final truth will turn out.

Correct. But you can examine someone's performance in her career up until now, and say with some confidence that she is an expect. She might lose her abilities and start making the wrong calls. People can lose expertise, for example as technology changes, or the market changes. The people at IBM were experts before 1980, but some of them lost their touch. The people now running Sears into the ground were once experts in retail, but they are no longer. Not in the new world of Amazon.com.

Also, you have to compare people to others in the same field. In basic scientific research, most people get it wrong most of the time. When someone said that Fleischmann and Pons were half wrong because of their neutron claims, Pons said "being half right is a good batting average in this business." (He said something like that -- I don't recall the exact words. He said it to me, and to others.)

In a war, a general who loses one battle but wins others is an expert, I suppose. In the hours following the attack on Pearl Harbor, headquarters in Washington informed every commanding officer of what happened, and warned them to watch out. Gen. MacArthur in the Philippines did nothing to prepare. His air forces were wiped out the next day. So you can argue he wasn't much of an expert at that stage in the war. Fortunately, he was a quick study and for most of the war he did an excellent job, as he himself pointed out on countless occasions.

http://www.pacificwar.org.au/Philippines/Japanattacks.html

Quote

MacArthur in the Philippines did nothing to prepare. His air forces were wiped out the next day. So you can argue he wasn't much of an expert at that stage in the war. Fortunately, he was a quick study and for most of the war he did an excellent job, as he himself pointed out on countless occasions.

Because he was ordered to do nothing by Roosevelt in the same way, like the Short and Kimmel admirals at Hawaii - that's known story. The USA needed an evasion for entering the war with Japan. How many examples of striking inactivity of otherwise experienced generals would you need?

Quote from Zephir_AWT

Because he was ordered to do nothing by Roosevelt in the same way, like the Short and Kimmel admirals at Hawaii - that's known story.

That's also bullshit conspiracy theory. The attack alone was more than enough to justify a war. It did not have to be a successful attack. If the U.S. had been prepared and if it had clobbered the Japanese fleet the first day, that would have all the better. There was no need to allow a surprise attack.

Quote from JedRothwell

"They debated new technologies, such as PCs, but the older executives who ran the company rarely dirtied their hands with the technologies enough to understand their potential and their problems. IBM had become like a music-publishing company run by deaf people."

JedRothwell :

During by practice work 1982 I visited many IBM groups later during research the IBM labs.

In fact IBM (and DEC too) was a split world. Famous research, sectarian top management. The code of conduct Watson (the IBM founder) wrote down, is pretty close to scientology logic and they lived it.

Usually IBM made all (relevant) deals over business (military, rotary, politics) connections, what helped them to survive even at the moment they completely lost the ground to DEC Alpha.

Just one example. 1986 they developed horizontal recording for hard drives, but they only sold high-end disks until a company called Quantum (later seagate – a spin of of DEC) killed their marked... The same happened for other technologies they invented. One exception: Josephson junction has been kidnapped by NORAD.

AlainCo. Interesting and Good points. You know I think where people see "conspiracy" it's quite rare that's it's really about money and organized behaviors far more often it comes down to a group with one of these views of consensus especially one of the last two groups. When consensus forms on the other side the result can be two polarized groups that see conspiracy on each other side and entrench themselves in fortresses of opposing views.

It's a shame as I think people are mostly much more diverse and interesting.

Using experts 'inexpertly' leads to policy failure